Posts tagged: llm Archive

If agents make prime a bit faster, what does that mean for the rest of us mortals?

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 AIs aren’t good rule followers

2026-04-14T15:16:11Z

Steve Yegge: I was chatting with my buddy at Google, who's been a tech director there for about 20 years, about their AI adoption. Craziest convo I've had all year. …

I’ve gotta agree with bob on this one, the first thing I did to my biggest brownfield project I wanted to use agents on BEFORE they did work was a hardened pre-commit.yaml, ci, hardened type checking and linting. SECOND get rid of bad inconsistent patterns, let them replicate consistency, force them to pass checks. Agents will follow all of your markdown suggestions most of the time, enough for you to become complacent if you let it. They are goal seeking, if you put them to a task you thought was possible that is not given your constraints, they will try to find a way given enough tokens. I dont see this ever changing, its one thing that makes them great, it just needs to be kept in check.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 A quote from Steve Yegge

2026-04-13T21:10:23Z

Steve Yegge

f91a8893-b1ba-422a-9390-18de5034483c.mp4

behind, yet positioned to completely dominate this race by hitting it with some sense. Making trends in what looks like longevity in the race that is not subsidising to simply get users, but to get by until they figure out how to 100x reduce the cost to a reasonable level. They feel like the guy sitting in the back with nothing big or flashy to say that is going to drop the hammer on their competition that overstretched itself taking on too much debt because it was necessary to change the game. There might be something to having a mix of hipsters, boomers, and luddites all trying to balance each other out.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Anthropic Mythos

2026-04-13T16:18:15Z

An ai model created by Anthropic was announced as a closed preview on April 7, 2026 for critical security research and evaluation with its close partners with critical software such as operating systems and browsers. Anthropic claims that mythos is able to reason through so much more context that any model ever before. This enables it to find bugs that are 25 years old in the BSD, considered one of the most secure operating systems we have. Once it finds these zero day bugs never discovered before its able to use them together in malicious ways never expected. In ways the world is not ready for. At the time of writing these are claims without proof. It remains scary to know the potential this has and that there is only a few companies with this potential that will gatekeep who gets access.

💭 How does Claude Code actually work? - YouTube

2026-04-13T10:15:26Z

5 star video, if you are going to watch one video to understand how harnesses and agents work, this is it. This really had my gears spinning on what tools do for agents and how big of a difference they make in their ability to manage context efficiently and accurately create changes. It’s crazy how good bash works, and that gives the agents the ability to do just about everything, but it could be better.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Agents Are Here

2026-04-10T16:11:37Z

🌱 This post is still growing

Late last year I started writing I'm Out On Agents. Agents sucked, the models were good, but there was still something missing between the harnesses and the models. They could write good code, they could do some debugging and exploring, but they were too good at fucking up the whole project to be useful. They could crank out Green Field POC’s like nobody’s business, but they created so much mess in brown field projects that it was easier to chat and edit yourself.

The Beautiful Glitch - Gemini

The Inflection Point #

It’s very well agreed on that the inflection point for most people happened with Anthropic Opus 4.5 in late Nov 2025. Early adopters probably noticed right away and shouted from the rooftops how good it was. But we’ve all heard that developers have 6 months before ai writes all the code for years, so this felt like the rest of the noise.

Hitting the December slowdown many of us hit code freezes at work. We completely disconnect from work for the last Week and come back in January. During this time, its very common for us to try out new tools, new techniques work on side projects, create POC for that thing we never have time for. While it looks like less features coming into the apps we support this is an important time for us to explore and reflect.

Anthropic was super generous with a free tier giving out huge levels of free tokens at the time. So many of us laughed and threw it at our side projects expecting the normal garbage output, but maybe some good ideas to come from it. But that’s not what was happening anymore, somehow these agents do some real work, follow plans and stick to scope really well. And if you lay out a big enough plan they tended to keep cooking and completing features.

January 2026 #

flu season

For me and many others around the country we were hit with a rough flu season, kicking in and draining us mentally for a good month or so. I wanted to work, I was getting excited about some projects and wanted to get them going, but I was constantly wiped and had no capacity. I had no ability to think on complex tasks, I was coughing all the time, just trying to survive, but yet wanted to do something. I started doing some small clean up, some work on side projects.

At this point I was still afraid of really letting agents cut loose on something meaningful, something that users depend on.

February 2026 #

what just happened?

yes or –dangerously-accept #

March 2026 #

April 2026 #

Closing the editor #

b289fda3-35e7-4b2f-aa21-6601c57d937a.mp4

Closing the Editor - Gemini

💭 An AI state of the union: We’ve passed the inflection point & ...

2026-04-08T09:29:40Z

A really interesting long form interview with @simonwillison.net. If you follow him closely most of it is probably not new, but I found some interesting nuggets.

Simon is writing most of his code from his phone these days using anthropic hosted platform. He mentioned that a lot of security risks go away when you don’t put secrets on the platform and you let them take the risk of running ai written code with ai chosen supply chain.

He talked about the Pelican Riding a Bike benchmark for quite awhile. He was surprised at how well of a proxy it is for how capable a model is at just about everything. He also said that when he runs the benchmark he also runs half a dozen others that he’s never talked about so that He could see if they were to train a model specific to his benchmark he could catch them, but it seems they had caught on and if they were they seem that they would already be doing it on all of his others anyways.

TDD is incredibly boring for humans, it strips so much creativity and joy from the process. Who cares if agents are bored they do better when doing TDD.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 @seldo.com on Bluesky

2026-04-08T07:31:34Z

Laurie Voss (@seldo.com)

Project Glasswing is a glimpse at an oncoming future in which agents do things humans could never have accomplished and the results are handled by other agents faster than humans could react and we...

Bluesky Social · bsky.app

Is Glasswing the next inflection point

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 What Happens When AI Stops Being Artificially Cheap | Daniel M...

2026-04-01T12:04:12Z

What Happens When AI Stops Being Artificially Cheap

The subsidy era is ending. Here

danielmiessler.com

I’ve been thinking about this for awhile and Daniel makes some great arguments here. Interestingly keeping inference cheap removes the incentives to make our tools better, help us choose the right model, lean on local models, open weight models. The frontier models are so affordable through subsidized subscription models why would you deal with anything less intelligent at this point. The tooling we use is not optimized for it, and why should it be.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 no one read the source

2026-03-31T18:47:59Z

Everyone look away, nothing to see here.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 safe words

2026-03-31T18:46:01Z

Anthropic safewords are the talk of the town today.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 you can replace the spinning verbs in Claude Code

2026-03-31T18:31:56Z

Nick Nisi (@nicknisi.com)

The claude code source code leaked today and the tweets are great, maybe twitter is back.

Did you know you can replace the spinning verbs in Claude Code. I’m having fun with it.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 @nicknisi.com on Bluesky

2026-03-31T10:19:49Z

Y'all, I think I'm a convert to pi

Bluesky Social · bsky.app

I’m about to be pi pilled.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 To Live In A World Without AI

2026-03-26T08:29:37Z

To Live In A World Without AI | Nic Payne

I'm finding lately that I wish we could go back to pre-ChatGPT... A world without a code-gen easy button, where "easy" was LSP autocomplete, wher

pype.dev

We f&#ing said @pype.dev, well f&#ing said. I think a lot of us are feeling this, we’ve pitched our brain into a bucket and we are no longer stretching it in the same way. We still work in similar ways of old, with new ways of turning off and saying yes a bunch of times. the best thing I can hope for is that as things get better we have fewer yes loops, and more architectural design debates and deep thoughts. But I fear deep thoughts are gone to the way of “research the leading 10 frameworks and pick the best one for this project.” and letting the clankers do the deep thinking. Its signing us up for a weird distopia.

I think a lot of us wish we could undo what has happened and go back to actually understanding what we are doing, but the world has changed, and if you are building average shit, like the average person, using models trained on average people doing average shit you cant keep up anymore.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 pype.dev - https://github.com/steveyegge/beads

2026-03-26T08:24:05Z

My Thoughts on Beads | Nic Payne

[Steve Yegge](https://en.wikipedia.org/wiki/Steve_Yegge) is a pretty well-known individual in the tech field, having been around for a long time at some of the

pype.dev

I’m in step with @pype.dev here, I really want beads to work for me, but my systems for infra/platform work are all over the place, not one repo. I’m considering trying the BEADS_DIR env var but idk if it fits my workflow. For now, similar to @pype.dev, I am rocking my own home vibed solution that I’ve intentionally put little effort in and its working great and I expect it to be broken and not working with the latest harnesses and models within a few months anyways, cause there is no predicting this train.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Has 3D Printing Reduced Adam Savage's Maker's High? - YouTube

2026-03-25T20:37:06Z

Vibe coding is going so far into the news sphere now that Adam Savage even weighs in with perspectives from someone who has built a life around building things with his hands, keeping up with new making techniques, discovering old techniques as they combine with new. He talks about 3d printing reviving his love of the pantograph as one automation technique eases the most difficult part of another.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Notes – 06:34 Mon 23 Mar 2026 – David Bushell – Web Dev (UK)

2026-03-23T09:07:38Z

Notes – 06:34 Mon 23 Mar 2026

dbushell.com · dbushell.com

Does anyone think fast-code will continue to pay the same salary? The answer isn’t to switch your brain off during your McCode shift and write a poem after work. Your job will be replaced by a Banglasdeshi slop-shop if AI improves (which is inevitable, apparently). Possibly the same sweatshop that loomed my £3 T-shirt. The Luddites didn’t accept their fate so easily.

David has some good points here, but I’m feeling the opposite direction a bit. Execs have always liked keeping the PM’s and the people steering the ship close by and were willing to farm out more and more grunt work. It feels like we are in a weird phase where there used to be a big group of people paid to write code. A few of them are exceptionally good at it and will remain. There will be a need for these people everywhere. Somehow we still need people hand editing assembly code optimizations, fortran, and cobol today. Those industries largely moved on, but a few great ones remain. I think this fast-code slop factory is going to be a short forgotten time in history, but no one yet knows what’s next. We are all waiting to find out. Just with anything there is still value in doing it by hand and understanding it at a deep level, but its quickly becoming hobby work. The things that most people code, things like CRUD apps do not require a special skill. They are largely replicating existing good patterns on frameworks to implement business logic that are custom to the app. There is no way of getting around the latter. Someone will always need to understand the business logic and be able to “sign off” and assert that it does the thing the business needs.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 The future of Coding and Code Quality - YouTube

2026-03-17T09:07:41Z

Very interesting takes from @thdxr in this interview. A lot has been hashed out by others all over the place, but a hot take here is that code quality is higher than ever right now. Codebases are becoming more consistent than ever. If you are not starting with a good consistent base from the start you are poising your context and doomed to fail and have all the common failures of ai written code. He still reads almost every PR, and will read all of the code eventually. There are a few cases where reading the PR is not worthwhile only when its low stakes, knows that good patterns have been established and followed. He argues that someone needs to be the expert of the code and of the product still and fears that too many people not looking at prs will fail companies.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Thinking about ai productivity again

2026-03-16T21:01:44Z

Thinking about AI productivity again. It’s allowing massive amounts of work to get done, to levels that humans cannot physically type out in some cases. But not all of this work is necessarily high value work. Right now I’m working on one of the biggest PRs to an internal cli library. Probably the largest PR I’ve ever done professionally. It touches all of the cli, refactors every command, reaches into the business logic layers to drive deeper separation. I reaches into the common layers to drive consistency. It ensures that every command (50 or so) has similar flags, supports –plain, –no-color. It specs out contracts to ensure that data goes out stdout, any extra goes out stderr. This makes everything unix pipe friendly. There was quite a bit of research and prep that went in, that turns out to already be distilled down into clig.dev. The point is that this is all good work. It will make the product consistent, repeatable, expected, and most of all boring. Most of the time, it will just work. Since we did it ahead of a lot of other agentic work on the product its establishing good patterns for the product moving forward. But its low value work. We wouldn’t have likely put humans on this work wholesale and fixed critical paths as they came up. Its not cutting cost, selling more product, or driving critical business decisions. Yes it’s worth it now, but it would not have bee in the past.

💭 Do You Have Token anxiety? - YouTube

2026-03-15T09:27:15Z

Kids are leaving the party early, not drinking, cant watch netflix without the laptop open. They are leaving the party early to check on their agents. I get it, that feeling that you need to eek out one more prompt, keep your agents running. if they arent running what are you even doing. If not you 6 others are ready to pass you up. The timeline to be first has shrunk to nothing but unachievable.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

We are the Grey Beards

2026-03-04T20:44:12Z

In November 2025 everyones beard lost its color, we aged into the next generation without realizing it. If you were getting paid to write code at this point in time, you are part of a special point in history where we used to write code by hand. There will be systems air gapped systems somewhere devs will continue to do it how we’ve always done it, some day they will peek out of this cave and realize that they are the only ones left, no one else remembers what its like. Writing code will quickly become a hobby that people do, in a weird niche way. Not because you want to build something, but like the guy with a mainframe in his garage that likes to watch the lights blink. Because its nostalgic, it’s a very cool skill, its fun and rewarding, but it won’t be to get something done.

Clankers got me tired

2026-03-03T20:42:25Z

I spent all day grinding on a 20 minute fix. I want the agents to do it. They can do it, but they are missing the harnesses they need to replicate my workflows of old.

💭 An attempt at a balanced perspective on AI - YouTube

2026-02-22T21:09:05Z

THIS, THIS is how most people are feeling about AI right now. Theres lots of “oh ai bad”, “but ai help”, “but ai company sleezy”. Cassidy did a fantastic job summarizing how most of us are feeling. Ending with well at the end of the day, I can’t do anything about the bad, the best thing I can do is learn how to embrace the good cause it aint going away any time soon.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Is Ai Faster Yet

2026-02-20T14:37:36Z

Is AI making us more productive yet, more faster yet?

Non-code #

probably not

I’ve seen this question hitting all over the Internet lately, and often points to people not writing code. Copilot turns prompts into emails, emails back into summaries that look a lot like prompts. I think there’s a place for this, making rambled thoughts sound more coherent, summarizing notes and meeting minutes. All good stuff but does it make us more productive, probably not by an amount that you can put $ $ behind, unless you are reducing headcount. thats not what we are doing right???

Coding #

with chat, probably not

When we talk about chatbots like gippity I think there’s a benefit to having someone with jr skills in everything to talk to, someone who can read all of the docs in an instant to get you some code snippet that might have taken all day to research and get right, but more productive, probably not.

Agentic Coding #

maybe

I’ve hit a stride with coding agents this year unlike I’ve hit before. Its able to bang out huge swaths of code in a prompt, with a good enough spec/plan. It absolutely churns out code. The bottleneck quickly becomes UAT.

I definitely feel like I’m doing a lot. There’s a lot that’s getting done that there’s no way I could do alone, it would take a full team with heavy coordination.

Am I more productive #

I’m definitely doing more, there are more lines of code, more ideas flowing, more docs written, more notes taken, more bugs squashed, more features out.

In the sense of generating more $ $ for the stakeholder. I really don’t know that I’m there yet, everything feels so new and different that there’s a bit of a learning curve to understand it all. There’s a pace and cadence to everything that may not be quite caught up yet.

It definitely feels like more is happening at a different pace that is hard to explain.

I’m sure there is a wide spectrum of folks out there from some generating millions of additional revenue in new greenfield industries, to those burning tokens and getting nothing in return.

With this SUPER AUTO COMPLETE engine it’s easy to say we found a bug affecting a user, but rather than squashing that one and getting a release out quickly squash ALL related bugs that a user might find in this category, it only takes a few more keystrokes right??

Right now it feels like a lot is happening but its hard to sus out the real productivity from the noise.

FIN #

As with anything AI It took me a few weeks to wrap this up and everything is already different. But I had some recent experiences of feeling slower than just doing it, and its all around context. I have built out a lot of the systems that I maintain by hand, I really know them well and I can solve a lot of problems within the hour by hand with coffee break. But now that I’m trying to lean on agents, I know they can do a more thorough job than I can faster, but I don’t have all of the data harnessed up. Just today I spent all day trying to solve CI problems. I can see the issue, I know the solution, but I want agents to solve this for me in the future so I spent the say working on a CLI to pull data from ci pipeline runs, manage the tokens get the logs, status, all of the data. Because I know this will be the way of the future, they will outpace me, I just need to have methods of getting context ready for them, and agents that know how to do it set up.

💭 POV: You install Clawdbot on VPS - YouTube

2026-02-16T10:22:40Z

Use a linux vps, It’s easy, just follow these simple instructions.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 OpenClaw: The Viral AI Agent that Broke the Internet - Peter S...

2026-02-16T10:19:08Z

Such a good interview @lexfridman is such a talented interview. It’s so cool to see the other side of this. For weeks we’ve heard about the story of the name change, we’ve seen everyone shitting on the security model, buying up all the mac minis in existance, fear mongering not to install this thing. @steipete.me has such a cool story from the beginning talking about making this thing fun and exciting. Giving it a personality that is not “You are absolutely right”. The story of changing the name twice, and getting pwnd on every step the first time and nailing it the second time is incredible. Dude is having fun trying to make the thing he wants in the world exist.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

How To Run 5 Agents In Parallel Feb 2026 Edition

2026-02-11T09:28:02Z

Are developers really running 5 agents in parallel? How the Heck do they keep up with the changes? This seems Impossible.

I was listening to Syntax.fm this morning and heard this question, and thought I’d throw in my take, which is probably pretty similar to Wes and Scott’s.

https://www.youtube.com/watch?v=NrBQI9So5lM&list=PLLnpHn493BHHNUfHN5lDf11UD8jQ5Bpzl&index=1&t=99s

Yes #

Yes, developers are running 5 agents in parallel. It’s not that hard it requires you to shift from thinking about the weeds and seeing the forest see: Pm Not Babysitter. It requires effort and diligence. Most importantly it requires planning, it requires agents, it requires tooling.

Is anyone doing this all day? Probably not. At least not outside of the startup companies that are building out tools to do this. Yes there are some, there’s always outliers, but its going to be rare. To have multiple agents running in parallel add day you need a lot of tokens, access to good models, and right now a low to medium risk application.

The big news right now is that Anthropic did what took google two years to do around 2008 (make a prototype browser render basic web pages) in 2 weeks. This took a swarm of agents running, a good plan (the modern browser is probably the most openly spec’d piece of software ever to exist, so they had a lot to go on). But this is also zero risk. If it doesn’t render there are no sales lost, no traffic accidents, no security breaches, no missed deadlines.

not with chat #

chatbots like chatgpt, are not getting you to run 5 agents in parallel. Maybe you take a sip of your coffee while it spits out its response, but you don’t have enough time to jump between many of them. If this is your experience so far, I understand how confusing it would be to think that someone is running 5 agents in parallel. But they are, and they are doing it with tools that are not chatbots.

PLANNING #

This is the core of what it takes to keep agents running for long periods of time in Feb 2026. Agents need something to do, telling them to turn the circle green, then blue, then to a rectangle, is not it, They will have this done in seconds. You probably could have done it just as fast and better.

For the plan itself, Send agents off to research. Agents are not good at solving problems without a good plan, but this does not mean you have to tediously write out the plan from scratch. They are really good a reading docs, specs, standards, finding them on the web. Reading your codebase to understand where and how a bug might be happening. They are really good at running bash, browser automation, they can reproduce your bugs for you and provide detailed logs with expected and actual behaviors in your plans.

Note

If you are trying to solve a production problem in an environment where you have risk and dont want agents running wild in, dont let them in, or scope their permissions. They are really good at understanding they cant access and writing a script for you, one that you can review to ensure they are gathering facts and not chaning production servers. Then you can run it yourself and paste the results back in. Sure there are better ways, but without any set up this works.

Executing plans, if you have well scoped and documented work for the agent to do, as of right now they are happy to keep working. I dont think there yet exists any sort of best practice here yet. I prefer to keep it out of the agent tools plan mode only because that is so deeply tied to the session. If you get context poisoning or a shit compaction your session might be hosed and unrecoverable, I’m sure you could go into the agents session files and do something to get it back, but I’ve yet to need one back that bad to really care.

What you need is a system that agents can access to tell them what to do, a simple markdown file works, GitHub issues work, I’m sure boards, and kanban work if it has an api the agent can understand.

Pm Not Babysitter

2026-02-06T14:04:25Z

Stop babysitting your agents, treat them like a real team and they will reward you.

Back in December I saw theo make a comment that code is now cheap, its the run rate of models, He quoted a study, not sure that he fully even believed it, but it claimed that the average developer after all meetings, training, emails, planning and extra shit in their day averages out 10 well tested lines of code per day. Opus 3.5 made him 10k loc (lines of code) that day.

We have all agreed for decades that lines of code is not a proxy to productivity or quality. Often more code means more risk, more review, more infrastructure. This has become MUCH different. Lines of code are still far from any sort of good metric. That aside, your agents are not doing 10k lines with you babysitting them, and in fact its very likely that the product quality is MUCH worse as you babysit them.

You need a tool for planning and tracking, otherwise you are playing babysitter rather than Product Manager (PM).

💭 The shovelware cometh

2026-02-03T21:45:51Z

The shovelware cometh

In September of last year, I covered a post by Mike Judge arguing that AI coding claims don’t add up, in which he asked this question:

If so many developers are so extraordinarily productive usi…

jerodsanto.net

Not surprising theirs a lag, between the models getting better, the tools getting better, and the masses getting better at using them, it takes time. This is still quite a hockey stick. I’m wondering how many are not posting on Show HN embarrassed they built something they know nothing about and afraid to get questions. I have no idea how anyone would get this ratio, but if I were a betting man, Id bet the ratio of build/show went way up. Plus we are probably getting a ton of people who have never heard of HN start building cool bespoke things for themselves and thats it, they use it, they love it, they might tell/show a friend.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 The shovelware cometh

2026-02-03T21:45:51Z

The shovelware cometh

In September of last year, I covered a post by Mike Judge arguing that AI coding claims don’t add up, in which he asked this question:

If so many developers are so extraordinarily productive usi…

jerodsanto.net

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 dax on X: "finally got around to setting up an always on openc...

2026-02-03T08:18:39Z

https://dropper.wayl.one/file/9065fcb2-5e40-479c-967e-498bc9bb6a4f.mp4

I tried this flow [of running an opencode server on tailscale] on day one of getting opencode, I wanted to prompt from my phone while were were running lights at the theater. It kinda worked, but the ui was really bad on phone, hard to use and the experience overall–it felt buggy. Happy to see they are making improvements and it might now be ready for some real use.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Ping 21

2026-01-29T07:30:59Z

Agents right now

can I access the project you mentioned?

yes

Can i access /tmp

yes, just do it

While I’m I at it, kubctl delete...

yanks plug front internet

Agent Management Is Exhausting

2026-01-26T14:38:35Z

The state of development in early 2026 is all wrapped around learning how to manage many agents running in parallel. Everyone’s trying to figure out the workflow.

The Plan Is Everything #

The secret I’ve discovered is a good, well-defined plan. This could be a markdown file or a GitHub issue. Agents are actually great at writing these for you. They’ll include reproduction steps, outline changes needed, and structure the work.

This is your opportunity to step in. Read the plan. Look for hallucinations. Spot where it’s going off track. Edit the plan before the agent starts coding.

I had one today where it laid out reproduction steps beautifully, but I could add context about network requests that completely changed the approach. This editing phase is what most people are missing right now. Skip it and you’ll watch your agent solve the wrong problem with impressive efficiency.

The Pace Problem #

Here’s what nobody warned me about: managing these things is exhausting.

Depending on the day, agents move so damn fast. I can barely research, find, and raise issues as fast as Claude can implement features and fixes. It’s like trying to speedrun a Minecraft seed when you just figured out how to craft a pickaxe.

A Different Kind of Work #

This stretches a different part of my brain than I’m used to using. I’m learning new skills around:

Issue tracking and management at high velocity
Knowing which models handle which tasks best
Spotting when a session is about to go sideways from context bloat or bad compaction

I had a session yesterday where the context got poisoned with a wrong assumption. The agent spent 20 minutes building on that false premise before I caught it. That’s 20 minutes of perfectly executed code solving the wrong problem entirely.

Worth It, But Hard #

I’m not going to hype-bro you and say this is easy or that you’re wrong to be skeptical. If you have the opportunity to work with agents, it’s worth learning proper techniques.

It’s not easy. It requires actual management skills, not just prompting skills. And yeah, I’m tired.

Stop Using Boomer Ai

2026-01-26T13:12:29Z

I was listening to these guys talk about migrating off of boomer ai the other day. Introducing the term boomer ai to describe using chat, copy, paste instead of agents. Something magical happened to the tooling and models around december, they got really good. The chatgpt $20 plan hooked into opencode is good, the Free models in Opencode Zen (Big Pickle and Kimi K2.5 Free) are really good. Neither of these quite match up to the speed and quality of the larger plans, but they are good. good enough to throw away your boomer ai techniques and start using agents. Agents are the future, and they are here now. If you are still using chat, copy, paste, you are doing it wrong. Stop using boomer ai and start using agents. You will be amazed at how much better your results will be.

https://www.youtube.com/watch?v=3dmPKuTWbsI

Dabbling In Go With Agents

2026-01-23T14:46:07Z

I’ve been dabbling with go this week using

What Your Coding Tool Says About You

2026-01-21T16:25:41Z

open code - libre free as in beer and speech
Copilot - corporate 9-5er
Cursor - You sip on Philz coffee with your macbook
Claude Code - Agentic Workflows or Bust
Jetbrains - I didn’t know you wrote java
Vim/neovim - definite neck beard
VsCode - What else is there?

Quick Tip To Get Agents Running Longer

2026-01-21T16:22:26Z

Here’s a quick tip if you want to get your agents running longer, prompt better.

💭 make no mistakes by Beginbot | Suno

2026-01-19T20:25:15Z

make no mistakes

Listen and make your own on Suno.

suno.com

Ai prompt music, the right level of cringe to be great.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 I Hope This Email Finds You Before I Do - Last Week in AWS Blog

2026-01-19T15:24:34Z

I Hope This Email Finds You Before I Do

Last Week in AWS · lastweekinaws.com

This post was filled with real life, snark, entertainment, feelings. I get a lot of these emails that claim they can change my SEO game if I give them 500, for a site making 0, Link partnerships from small startups. A lot of these are so automated that if I do respond I dont even get a response. I’ve responded to many simply asking what is this about, I get 10 others just like you a week. Can you tell me what exactly you want and what each of us benefit from it, how did you find me. All normal questions, they almost always result in crickets, no response, maybe its time to implement a Billie for more snark.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 "I shipped code I don't understand and I bet you have too" – J...

2026-01-14T02:15:41Z

Absolute banger of a conference talk, saving to come back with more notes later.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

opencode variants

2026-01-13T08:53:36Z

Opencode is changing on the daily right now, today I noticed the word low pop up in Orange text in my opencode window. Looking into this they are exposing variants to the user. This allows you to change between fast or slow and thinking, the later taking more time to prepare before doing an action.

Swapping Variants in opencode

It looks like this toggle may have been here for awhile and I’m just now discovering it. Potentially because its a new feature of the free Zen provider.

Variants Many models support multiple variants with different configurations. OpenCode ships with built-in default variants for popular providers.

Built-in variants OpenCode ships with default variants for many providers:

Anthropic:

high - High thinking budget (default) max - Maximum thinking budget OpenAI:

Varies by model but roughly:

none - No reasoning minimal - Minimal reasoning effort low - Low reasoning effort medium - Medium reasoning effort high - High reasoning effort xhigh - Extra high reasoning effort Google:

low - Lower effort/token budget high - Higher effort/token budget

Context Is King

2026-01-13T08:28:23Z

A new approach to agentic workflows.

This is probably news to no one else, I’m sure I’m behind on this one. You can’t one sentence prompt and expect to get what you want.

Agents.md #

Writing Complete Prompts #

Ping 17

2026-01-12T12:01:16Z

I’m not sure if its good to keep issues and plans in the repo long term, but I’m going to give it a shot.

read @pages/issues/issue-1.md create a plan to fix the issue, then write your plan to fix in @pages/plans/fix-issue-1.md

💭 Don't fall into the anti-AI hype

2026-01-12T03:03:25Z

Don’t fall into the anti-AI hype

I'm glad someone was brave enough to say this. There is a lot of anti-AI sentiment in the software development community these days. Much of it is justified, but if …

In the near future, transforming text over an entire document will become as commonplace as filtering images.

Never believe in absolutes, see what I did there. The hype bros will take you to the extremes, ai will take your jobs in six months or be burned to the ground in six months. How about its useful now and will be more useful in six months. If you turned off the hype bro feed for six months you would probably be fine, in fact you would probably be better off for not capturing so much noise along the way. AI has gone the way of next js framework, it churns fast, hype bros are always an expert that know exactly whats best for everyone. It changes fast, what was the best last week might be dead next week. In fact getting to know what works well for you and knowing that tool really well for a longer period will take you farther.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Salesforces mistake - YouTube

2026-01-12T02:52:55Z

Salesforce gets pwnd by the ai hype bros and killed its reputation with employees, letting them know where they truely stand with them. 4k people sounds like a lot, its probably a big chunk of savings, but was it worth the loss of reputation? There must be a better way to give this a trial run that lets them understand this before disrupting the lives of real people right???

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

Ai

2026-01-11T13:48:43Z

Last updated Jan 2026.

Seealso

Looking for all posts tagged with ai see [[ tag/ai ]]

Ai is a tool I use a lot for code generation, research, image generation, and debugging. The words I publish on this site are my own unless explicitly stated from the top. There’s only one or two posts in this category.

Words are mine #

The core of what this blog is, is my thoughts ideas, sharing experiences. The words are the important part. They are not perfect, I often do not spell or grammer check, and what is here is from a flow state of writing and very often not refactored.

No AI unless explicitly stated

code gen #

Up till around 2023 all of the code to create the site was hand written by me. I have my own site generator that I maintain markata.

From 2023 through 2025 snippets of code, sometimes entire plugins or modules were created in chat apps like chatgpt. There was no agentic coding, windsurf, cursor, claude code, opencode, none of it. It was all integrated into the site by hand.

Starting around Dec 2025 I was finding freely available agents through tools like opencode to be more and more useful, making less mistakes, doing better at getting context and making changes. The flow is relying on heavy planning and issue reporting.

image gen #

I really got into stable diffusion as it came out. I thought it was fascinating to generate different kinds of images of things that I would otherwise have no ability to create, for a long time I was generating images for every post. I’ve moved on from doing it for every post, but still do it occasionally.

I’m not an artist, sometimes I like to include art that I would not otherwise be able to. Most of the art you see on a post is AI generated, and may not be called out.

Techbrophobic #

I feel this term strongly. I find ai tooling useful, but the industry and hype bros bring a bad taste.

The tools do not not do everything right, you still have to think, watch its output. You have to put the work in to do proper software engineering. Its not a magic pill, but it has the ability to implement ideas, plans, very quickly.

Also the whole industry creeps me out. The theft that happened to get here was wrong. AI shovoled into everything as slop sucks. The stock market pump is sus.

I'm In On Agents

2026-01-11T12:02:15Z

It’s the start of 2026 and agents are getting a lot better than they were. I’m using opencode at home, free mode with Zen and big pickle. At work I have access to a wider variety of models including what seems to be the gold standard 3 from anthropic opus, sonnet, haiku.

Note

I’ve sat on this post for over a month, refactored it a few times, things feel like they are moving so fast that its hard to keep my feelings straight.

Things are changing #

Around Aug 2025 I wrote I'm Out On Agents. I saw others in the space having such great success I gave it a solid shot, but found it to egregious edit more than I asked, make massive unneeded changes, and make more small bugs hidden in the details than was worth it.

This was just after the tipping point where not just hype bros were making the switch, but people I work with, people I trust and have made really good Software in their career. People like the creator of Flask Armin Ronacher. People who are really good at writing code and do not need it to do the work were finding themselves more efficient working with agents than making edits by hand.

Planning #

Currently as I see it there is no free lunch. Software engineering is still very much needed, but the work is switching. The models have been good for awhile, they don’t seem like they are getting leaps better with each release like they were for so long, no matter how much t3.gg hypes up the new models. The tools feel like what is really making the difference right now. The ability to take short human prompts figure out what context is needed, getting that context, and keeping it in the context window is getting better and better making agents useful for more things.

These improvements in tooling is making our job look different. Knowing syntax and how to write clean efficient DRY code is less and less of the job every day. Learning how to create detailed plans with the help of agents, editing them, knowing when they are good enough that you can be confident in an agents ability to execute on them. Knowing how to balance the context is becoming the job. Moving forward its likely that models can handle more context and this will look different, and change, but planning and context management is a key part of the job moving forward.

Context is king, good plans are paramount, syntax barely matter.

Issues #

💭 Oxide and Friends 1/5/2026 -- Predictions 2026!! - YouTube

2026-01-09T21:41:43Z

I thought this was an interesting take from Simon. I’ve been hearing him consistently say there will be more demand for software engineering in the future. More companies will have the ability and need to deal with software applications, but fewer of us will be hand editing any code. I thought this was an interesting interaction in the clip.

Uh Simon, what do you got for us?

I’ve just got the one. I think the act

of the the the job of being paid money to type code into a computer Yeah.

will go the same way as punching punch cards.

Okay.

I do I think in

six years time I do not think it will anyone will be paid to 80:56 just do the thing where you type the code.

Just type the code. Okay.

I think software engineering will still be an enormous career. I just think the software engineers won’t be spending multiple hours of their day in a text editor typing out syntax. 81:09

It will look like punching cards. I think so. Yeah.

Yeah. Interesting. In uh in six years. Um and but software engineering still very much exists.

I believe so. I I hope so. [laughter] I very much hope so because I think

the um the challenge of being a software 81:23 engineer is not remembering

how if what what if what for loops look like. It is understanding what computers can do and how to turn fuzzy human requirements into actual like working software. And that’s that’s what we’re for. And I think we’ll still be doing that just a lot more of it in a lot more ambitious scale.

And then Okay. Do you Does the software engineer though deals with code? I mean they I mean the code is being written.

I think they probably look at it occasionally.

Okay. Only occasionally a little bit. So I met 81:53

Who debugs it?

Um I I hate to say it, the agents debug it themselves.

Okay. Who debugs your device driver that either works or doesn’t? I like

working on this programming language like I’m doing my own codegen and like

claude is happy to pull out GDB and just like

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

opencode init prompt

2026-01-09T13:32:12Z

This is opencode’s init prompt.

Please analyze this codebase and create an AGENTS.md file containing:
1. Build/lint/test commands - especially for running a single test
2. Code style guidelines including imports, formatting, types, naming conventions, error handling, etc.
The file you create will be given to agentic coding agents (such as yourself) that operate in this repository. Make it about 150 lines long.
If there are Cursor rules (in .cursor/rules/ or .cursorrules) or Copilot rules (in .github/copilot-instructions.md), make sure to include them.
If there's already an AGENTS.md, improve it if it's located in <dir>

💭 Photoshop for text — Steph Ango

2026-01-06T02:48:34Z

Photoshop for text

Steph Ango · stephango.com

While the non deterministic nature of llms scare the heck out of me in the sense of just cutting it loose on my writing. letting it go through all of my files and just edit them. I do like the idea of mundane tools like “desaturate”, “Gaussian blur”, evolving out of it for text. I don’t yet see this with the tools we have now, but it will be interesting to see them evolve.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Vibe code is legacy code | Val Town Blog

2025-09-03T02:14:45Z

Vibe code is legacy code

Updates and articles from the Val Town team

blog.val.town

Brilliantly said. Vibe coding is legacy code. It’s code that we forget exists. Code that no one touches, you replace it. If you touch it you are more likely to break it.

The worst possible situation is to have a non-programmer vibe code a large project that they intend to maintain. This would be the equivalent of giving a credit card to a child without first explaining the concept of debt.

As you can imagine, the first phase is ecstatic. I can wave this little piece of plastic in stores and take whatever I want! …

When To Vibe Code

2025-09-02T19:44:20Z

I enjoyed this post from Theo and think it deserves re-iterated, revisited, and to remind myself of some of these things.

https://youtu.be/6TMPWvPG5GA?si=guQem4R8dLOMBntP&t=1356

The skill/read spectrum #

The first diagram describes that there has become a spectrum of agentic coding from vibe coding where you don’t ready anything, to looking at everything in detail, across a group of people who don’t have a clue what the code says to people who could do it way better if they took the time.

The importance spectrum #

He argues here that its ok to bounce between A,B, and D, but C becomes dangerous. I’d argue that he brought up a 3rd spectrum that is important later on, “how critical is this”. I think the I don’t know, don’t care, didn’t read, but the thing did its job is quite fine, but don’t know, don’t care, mission critical is the main issue we are seeing with agentic coding, primarily in the didn’t read but critical Zone.

The Rules #

This is the list that prompted this post, I think it serves as a good reminder when you should care a bit more.

You still need to know how code works if you want to be a coder #

If you want to get good, you need to put in the reps, do the practice, learn to debug the strange looking error messages and not just pass them to ChatGPT.

There’s a lot of code worth having that is not worth writing or reading #

As professionals we have a lot of code that we read more than we write, will get executed millions of times over the next 10 years that its in service, pay close attention to this and probably set aside the llm.

We also write a lot of that is ran only a few times, maybe its a special report, or a shell script to bootstrap something. It might be a POC service that you build out only ever on your machine, you get your idea out in front of users or try it yourself and find the mistakes before you spend weeks building it yourself.

you can’t be mad at vibe coding AND be mad at left-pad #

This one is interesting that he was very strongly for. The idea is that no one ever reads the code from libraries they use, or the diffs of new versions (outside of google where everything is literally vendored into the monolith). If you don’t care about all of those dependencies, you cannot get mad at vibe coding.

I’ll argue here a smidge, but I agree with the premiss. Left-pad was probably written by someone smarter than average. It was definitely used by far more people and projects, therefore was battle tested to no end. This was the real reason this stuff got in everywhere. People are probably not good enough to catch all the odd edge cases for js, especially if front end is not their specialty.

Vibe code is a type of legacy code #

Vibe coding is a type of legacy code, but as a type of debt that we opt into, not one that has accumulated over time and we have opted not to take care of. Both are types of code that no one reads, no one remembers existed.

When someone needs to change legacy code they don’t, they take a hammer to it and rewrite it from scratch.

If the tools are better than you, stop using them #

Theo has a good example here. Critial business components to his businesses have switched to using Effect. It solves async issues with ease, but melts the brain of anyone seeing it for the first time. He admits that he does not code as much anymore and thus he sucks at it, but knows that if he is ever to learn it one bit he must turn off the LLM and put in the work to do it himself.

You will not get any better at this by copy pasting from ChatGPT. It is too easy to put your brain aside, paste error messages, and copy the reponse in. Or worse with agents tell them fix this. Unlike the days of Stack Overflow, you had to understand your problem well enough to search it. You had to understand the answers enough to integrate the solution, you had to understand if the solution was even for your problem in the first place. There was much more brain work that had to happen even then when the answers were basically given to you.

Fin #

Lastly Theo does not bring this rule, up, but if you are writing text for humans to read, you shall not copy paste from ChatGPT without thoroughly editing and reading for yourself first, this is considered Ai Slop and you should be ashamed.

I'm Out On Agents

2025-08-27T10:10:51Z

Its the year 2025 and we are only a few years into having 6 months to live before ai takes our jobs, and the big push right now is agents, managing agents. I will fully concede to I’m not doing it right, or a future state gets better than where we are right now, but right now they kinda suck.

Transparency

I’m sitting offline right now as I write this, These are my feels, no research, no links, no ai, just vibes.

Chat #

Chat is what really kicked off ai uses and goes back as old as computers, but it always sucked. Then chatgpt rocked the world with the biggest launch day in history and showed us that it could actually be pretty good. Unethically trained on everything they could get their hands on, burning cities worth of electricity to train, and keep training to stay ahead of the competition. It does a damn good job. There are tells, and if you see enough of it there is a lot that turns to slop, but if you had never seen it before, there is no way you would assume that it was not a computer.

It does a damn good job at being average, it can do what seems like everything not related to security and authentication at a pretty average level. This is its super power. Whenever you are trying to bridge between something you know and something new, you can get a pretty good answer, and likely spot the bs in what you know.

Agents #

Now that our models have gotten better, hardware has gotten bigger, better, and a lot more of it, we can really expand context windows really wide. With that brought the use of agents, these tools get context from sources on their own and often are given read/write access to your computer. Depending on the model these things will branch out to make small changes that look no worse than a formatter on every goddam line of your codebase. Except they are not a formatter, they are not backed by ast checks. They do not have any guarantees.

They take the fun out of creation.

The emphasis now becomes on the code review. All you have to do is ask it to makes changes for you. Bring in your expertise of what changes should be next, or even just punt to asking it what comes next. I’ve yet to talk to someone that is diligent enough to read everything it spits out in excruciating detail to the point that it does not cause significant issues.

You see here is the thing, its average. With a little bit of context it can do average work. It is not an expert. On the surface this feels fine, making crud endpoints has been done for decades, and average developers crush these things every day no problem.

It’s Average at best

You know what is not average? Your knowledge of the use case you are solving. You may not feel like an expert, but given that there are likely about 5 people working in your codebase, you are a fukin expert at that thing. The average person off the street takes time to onboard, often months, or years for someone to really understand the business you are working in. These things don’t have that.

I’m not letting it in shit that I care avout #

I was early to the game to codeium, even used the predecessor for awhile, I was early to chat gpt, I was early to windsurf and the possibilities that agentic ides brought. I will be late on letting agents touch my production code bases. The few times I’ve tried for changes that seem easy enough, but more work than what I want to do at the time, I’ve regretted it. It’s only left behind a mess that it cant deal with anymore, runs in circles trying to solve any problem, and I’m left with shit that feels like a house of cards that breaks anytime you touch it.

I will continue to POC #

As much of a mess as these things, make they are super useful to vibe code ideas, move quickly and try different approaches to a problem. They let you make a proof of concept that you can get in front of team mates, bosses, or users.

Honestly I still prefer the chat interface. It feels like a nice balance of using my brain, and knowing that I am punting on something. Do I need to know the whole ffmpeg interface to grab a thumbnail for my webapp, no, would I actually like to learn it someday, yes, but I don’t need to right now I just need a goddamn thumbnail.

For the most part I am still slotting these things into the codebase myself, but occasionally I have it do an entire module, and even more rare I pull out agents and have it do all the work. The further away from the code I get, the less I care about it and just want it done.

I will laugh at this post in 2 years #

The way I see it agentic coding is not here to stay, it sucks. There is one two ways to go this was a blip in the radar that we laugh at for giving ai companies all our data for free so they could fuck up our products. Or they actually get pretty good and we all become architects that peek at the code if we really need to. I think for the second to become true its going to take a lot of time, consumer hardware will need to catch up, local llms will need to work a lot better than they do now. Even if we still need to call out to the big boys for some heavy work getting the context right for them locally would make a huge difference. Currently local llms are too slow and dumb.

So will this bubble pop or explode, we can only wait to find out.

💭 The Death of the User Interface

2025-08-25T01:48:19Z

THE_DEATH_OF_THE_USER_INTERFACE.md markdown

# The Death of the User Interface

> **TL;DR:** We're witnessing the end of graphical user interfaces. AI agents like Claude Code are eliminating the need for windows, menus, and clicks, replacing them with natural language. The computer is finally learning to speak human, not the other way around.

---

## 🔮 A Personal Revelation

Last week, I realized something profound: **I haven't opened Finder in months.** Not once.

Where I once clicked through nested folders, dragged and dropped files, and navigated hierarchical menus, I now simply tell Claude Code exactly what I need:

- _"Find all the test files modified in the last week"_
- _"Move the old backups to archive"_

The commands execute instantly, precisely, without me ever seeing a window, icon, or folder.

> This isn't just about convenience. It's a fundamental shift in how humans interact with computers, and it signals the beginning of the end for user interfaces as we know them.

---

## 🚴 → 🚀 The Bicycle That Became a Teleporter

In 1990, Steve Jobs famously described computers as "bicycles for the mind," drawing from a Scientific American study showing that humans on bicycles were the most efficient locomotors on Earth. The metaphor was perfect for its time: computers amplified human cognitive abilities just as bicycles amplified our physical capabilities.

But bicycles still require you to:

- **Pedal** the mechanism
- **Steer** the direction
- **Navigate** the terrain
- **Learn** the balance

Traditional user interfaces work the same way. They're tools that amplify our abilities, but only after we learn their language, their layouts, their logic.

> **What we have now with AI agents isn't a bicycle anymore. It's a teleporter.** You simply state your destination, and you arrive.

---

## 📜 From Xerox PARC to Natural Language: A 50-Year Arc

### The Timeline of Interface Evolution

**1964** → Douglas Engelbart invents the computer mouse at Stanford Research Institute

**1973** → Xerox PARC develops the Alto, the first computer with a GUI

**1979** → Steve Jobs sees the Alto, immediately grasps its revolutionary potential

**1984** → Macintosh launches, bringing GUI to the masses

**2024** → AI agents begin replacing graphical interfaces entirely

That language dominated for five decades. Windows, Mac OS, and even modern web applications all speak variations of it: _point, click, drag, drop, menu, submenu, dialog box, button._ We became so fluent in this language that we forgot it was a language at all.

### The Abstraction Layer Pattern

Every abstraction layer in computing eventually gets replaced by a higher-level one:

| **Era** | **From**            | **To**                             |
| ------- | ------------------- | ---------------------------------- |
| 1950s   | Machine code        | → Assembly language                |
| 1960s   | Assembly            | → High-level programming languages |
| 1980s   | Command line        | → Graphical user interfaces        |
| 2000s   | Native apps         | → Web applications                 |
| 2020s   | **User interfaces** | **→ Conversational AI agents**     |

> Each transition follows the same pattern: what once required specialized knowledge becomes accessible through more natural, intuitive interaction.

---

## 👻 The Invisible Operating System

Traditional operating systems: Windows, macOS, Linux, are abstractions over hardware. Web applications are abstractions over REST APIs. Both require user interfaces because they need to translate between human intent and machine execution.

**AI agents represent something fundamentally different:** they're abstractions that understand human intent directly. No translation required.

### Consider the Mental Journey of a Simple Task

🖱️ Traditional UI Approach

1. Open Finder/Explorer _(remember where it is)_
2. Navigate to directory _(remember the path)_
3. Scan through files _(parse visual information)_
4. Select multiple files _(remember shortcuts)_
5. Right-click for menu _(know this exists)_
6. Choose "Move to..." _(understand terminology)_
7. Navigate to destination _(remember another path)_
8. Confirm operation _(hope you got it right)_

🗣️ AI Agent Approach

1. "Move all PDF files from Downloads to Documents/Reports"

**Done.**

> The difference isn't just efficiency, it's cognitive load. With traditional interfaces, you're translating your intent into the computer's language. With AI agents, the computer learns your language instead.

---

## 🧠 The Mental Load Revolution

Every interface element, every button, menu, icon, and widget, is a **tiny cognitive tax**. Even the most intuitive interface requires you to:

- ✓ Understand its visual language
- ✓ Remember its organizational structure
- ✓ Learn its interaction patterns
- ✓ Maintain mental models of its state

This is what UX designers call **"extraneous cognitive load"**. Mental effort spent on using the tool rather than accomplishing the task.

> When you tell Claude Code to "set up a new Python project with pytest and black pre-configured," you're expressing pure intent. The mental energy you would have spent on navigation can be redirected to actual problem-solving.

---

## ⚡ The Present: Early Adopters and Edge Cases

We're living through the transition right now.

### What's Happening in 2024

- **AIOS** → Embedding LLMs directly into operating systems
- **Claude Code** → Replacing entire categories of developer tools
- **Cursor & Copilot** → Making IDEs conversational
- **Warp Agent Mode** → LLMs in the terminal for multi-step workflows

### What I No Longer Do

I see it in my own work every day. I no longer:

❌ Browse through file explorers  
❌ Click through git GUIs  
❌ Navigate package manager interfaces  
❌ Hunt through documentation sites  
❌ Configure tools through preference panes

Instead, I describe what I want, and it happens. **The interface hasn't been simplified, it's been eliminated.**

---

## 🍎 The Future Steve Jobs Glimpsed

> "Ultimately computers are going to be a tool for communication. Not computation, not productivity. Communication."
>
> — Steve Jobs, 1983 International Design Conference

At that conference in Aspen, a 28-year-old Jobs made predictions that seemed like science fiction:

- Portable computers with wireless connections
- Instant access to remote databases
- Devices as primary means of communication

He was right about all of it, but even his vision was constrained by the paradigm of his time. He imagined better interfaces, more intuitive interactions, simpler designs.

**He couldn't imagine no interface at all.**

Yet in that quote above, Jobs understood something fundamental: the real revolution would come when computers could understand us as naturally as we understand each other.

> That future is arriving. The question isn't whether AI will replace user interfaces, but how quickly and how completely.

---

## 🔄 The Last Interface

There's an irony in writing about the death of user interfaces, or rather, there **was**. This article itself is proof of the transition: generated through conversation with Claude Code, shaped by human intent rather than human interface manipulation. I provided the ideas and direction; the AI handled the execution. **The future isn't coming, it's already here, manifesting through the very words you're reading.**

Soon, articles like this won't be "written" in the traditional sense. They'll be conversed into existence, with AI agents handling not just the typing but the research, fact-checking, formatting, and publishing. The tool will disappear into the task.

### The Holdouts and the Inevitable

Some will mourn this loss. There's something satisfying about direct manipulation, about seeing and controlling every step. Just as some still prefer command lines to GUIs, some will always prefer clicking to conversing.

But for most of us, the appeal of **zero cognitive load** will be irresistible.

> Why learn an interface when you can just say what you want?  
> Why navigate when you can simply arrive?

---

## 🎯 Conclusion: After the Interface

We stand at an inflection point. For fifty years, ever since Xerox PARC invented the GUI, we've been refining the same basic paradigm: **humans learning to speak computer**.

Now, **computers are learning to speak human**.

The death of the user interface doesn't mean the death of design or user experience. If anything, it makes them more important. When the interface disappears, what remains is pure interaction design: understanding human intent, anticipating needs, handling edge cases gracefully.

The challenge shifts from:

- ❌ _"How do we make this button more obvious?"_
- ✅ **"How do we understand what the user really wants?"**

> Steve Jobs gave us bicycles for the mind.  
> AI agents are giving us something else entirely: **minds that understand our minds.**  
> No pedaling required.

**The user interface is dying, and that's the most user-friendly thing that could possibly happen.**

---

_What do you think? Are we witnessing the end of user interfaces, or just another evolution? How has AI changed your own relationship with traditional software interfaces?_

This is an insane level of agentic llm use, the author claims to not even use his filesystem anymore, its too cumbersome to find where downloads and documents are and way too easy to ask an agent to move all pdf’s from downloads to documents.

This scares me on multiple levels, theres the epidemic of ai datacenters and ai companies burning cash, burning through gpu’s and api calls to a giant data center just to move files sounds absolutely insane to me.

Then there is the level of accuracy. There’s a level of interpretation that happens with english that does not happen in code, code generally does what it does repeatably. The examples of “Move old backups to archive” is so open for interpretation that its ready to ruin your day, which backup, which archive, how old? Dude is ready to loose his files.

Last is privacy, there is going to be a privacy epidemic that is going to pwn so many people giving these things full access to email, chat, ALL of YOUR FILES, and computer, your whole network.

Look I get it this is where things are going, but dude is early, like too early. These things need to become far more accurate, less power hungry, and run locally before I give up my whole filesystem.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

A More Human Internet

2025-08-21T14:36:00Z

An idea for a post that may never happen

💭 The Internet is Dying: AI, Bots, and The End of Human Content ...

2025-08-18T13:04:56Z

Damn this VAnessa is hitting my feed with hard topics, I’m not sure whether to subscribe or to block. These top websites only feel worse every day, when I post on twitter and I get 4 likes by accounts that were created 5 minutes ago with racy profile pics it feels obvious. I wonder how larger accounts deal with it. Now that llms have made making these bots mimic humans easy It really makes you want out. I’ve really become a curmudgeon and leaning on rss over the past year, I dont like it, but idk what to do.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 The Real Reason Your Power Bill Doubled (It's Not Inflation) -...

2025-08-18T13:01:04Z

This is a crazy theory I did not realize was out there, but damn power just keeps costing more and more. She does not mention it here, but there are many sources of power for the grid that cost vastly different amounts to produce, generally “clean energy” solutions are harder and more expensive to bring online and don’t just turn on and off at the flick of a switch. Anyways, how are the power companies divying this power out to users, do some get preferred rates or supply? My rates just went up for the summer period “temporarily”. Our infrastructure is aging hard to upgrade and needs something done to it. Who’s really going to pay for it, these AI companies are throwing big numbers but do they have any real money? Do they have any real revenue after building out massive data centers filling them with the most expensive hardware? These guys are burning cash like crazy.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Slops | justin․searls․co

2025-08-10T17:41:56Z

Slops

AI-generated slop that I thought was worth sharing.

justin․searls․co · justin.searls.co

Justin has such great feeds on his site, I love how the main feeds are so prominant just to the left of the article you are reading. slops in particular feels like a great category. Saving this chat for later, or found it particularly interesting, but don’t really want to make a post about it.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 ThePrimeagen on X: "there is literally no universe that this i...

2025-08-09T00:09:36Z

there is literally no universe that this is true 10k lines and its not bug filled crap? ok Lex Luthor, its time to step away from the keys

Is this 10k real production code? Dry in the sense that it hasn’t re-implemented the same s3 api dozens of time? What language are we talking something dense like python? something very verbose like html? Maybe a language where you implement everything from scratch like lua. This matters a lot. Playing with little POC applications that dont mean anything I can quickly come up with 500-1k likes of code that I may never look at again. I’m sure I can come up wtih 10k decent lines of code a day.

But for the same application without duplicating everything over and over? For something that moves the needle and really matters?? every single day?? Consistently +10k, not 10k changes, not 10k deletes of yesterdays code. nah thats wack.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 I Will Fucking Piledrive You If You Mention AI Again — Ludicity

2025-08-06T16:39:28Z

Fantastic write up on their experience in ai, opinions on ai being a hoax with a veil of reasonable usefulness. Arguing that most people do not understand enough to see the difference, and thought leaders see where it is now, see where it was yesterday, it must be going to general intelligence tomorrow and you all will loose your jobs without this. I appreciate the satirical language here.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Meet Gor | I kind of hate Agentic IDEs for the sake of product...

2025-08-02T15:00:38Z

What ChatGPT is NOT - Tech Raven Blog

meetgor.com

Yes, I can review the code and make changes, but who in the world loves reviewing code? Do you love reviewing peers’ PRs? really?

I’m with MeetGor here 100%. reviewing the nuance, not being as involved with the process of creating the architecture design, not solving the problems that arise in development make it hard to effectively review and not turn into LGTM man.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

using yek to serialize text files into llm friendly file

2025-06-11T11:01:34Z

I’ve been using gitingest web ui [[ thoughts-516 ]] for quite awhile to serialize git repo into llm friendly text files. This gives tools context about repos that are not in the training data so that it knows about it and how to use the code in the repo. gitingest also has a python library [[ thoughts-517 ]]

I had a use case for a project not yet on git, and found yek.

Installing yek #

Their instructions tell you to curl to bash.

curl -fsSL https://bodo.run/yek.sh | bash

I don’t like curl to bash from random sites, so I have my own self hosted version of i.jpillora.com. I like using this because it pulls from github and I trust github as a source for artifacts as good as the repo I am pulling from.

curl https://i.jpillora.com/bodo-run/yek | bash

Using yek #

yek

/tmp/yek-output/yek-output-bb01e621.txt

This will give you a link to a text file that you can add to many llm tools. This happened so fast for me that I didn’t even believe that it worked properly.

more options #

As with most clis, you can run yek --help to see the options available.

yek --help

💭 What ChatGPT is NOT - Tech Raven Blog

2025-05-09T13:45:18Z

There is a lot of excitement about ChatGPT and how it allows us to interact with information and technology. I am actually excited that it now exists and still, I think it is being way overhyped. I...

Tech Raven Blog - · blog.techravenconsulting.com

Do you remember regression models from college: given some data, you find a best fit line that allows you to predict Y given X. At the end of the day, ChatGPT, and LLMs in general, are the same thing as the regression model – it’s just that ChatGPT is the largest and fanciest model we currently have to model language and information.

I really am coming to the idea of calling it a “word calculator”, this seems to be the most succinct description of llms that the lay person can comprehend and relate to.

ChatGPT does not hallucinate or become unhinged

I think Steve goes much deeper on this in his intervew on fafo.fm. They describe it more as a pleaser or “yes man” essentially all the companies that are building these models want to give the “best” answer, better than their competitors. With this comes the risk of it being completely wrong, they are designed to always give an answer.

On that episode they talk about how each word has a statistical probability tied to it and in some visualizations you can see the confidence, in almost every example of “hallucination” you will find some set of very poor confidence followed by BS. Thats not to say that every time there is low confidence there is BS though, that is why this is such a hard problem.

ChatGPT is not a replacement for search

I’ve become really bad at this. This is slowly becoming my starting point for a lot of research, not my only research, but the first wet finger in the air. Most often done with low effort, low risk tasks.

In ChatGPT, just like in the shelf full of the Encyclopedia Britannica I had when I was a kid, information on topics has been reduced to good summaries and presented in logical format. In this case we have removed the human authors and replaced them with a very large neural network.

I really like this description and it matches up with my last comment.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 The Rule - YouTube

2025-05-09T13:01:16Z

Under 2000 everything is happy, green field. Any decision you have made is relatively easy to back out of (barring you making a library with downstream users), but as you go, regret kicks in. Regret we didn’t make that pydantic 2 upgrade earlier, as new features become more apealing. Regret that we chose sqlite for simplicity, speed, agility, and now we might need robust and distributed. Regret that you chose a front end framework, or to have a front end at all to a backend problem. Regret that you put 6 layers of abstraction on your db early on and now that you understand the problem you want different abstractions, but all of your endpoints deeply depend on the current one.

Vibe coding will not save you, it will only make these wrong decisions for you without the context that you have. You will hate it’s decisions more because you had no input into some of them.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Vectorizing Your Databases with Steve Pousty

2025-05-04T17:54:39Z

Vectorizing Your Databases with Steve Pousty

What exactly is an LLM doing and why do you need to learn so many new terms? Steve Pousty is here to explain that most of those new terms are things you already kno…

Fork Around And Find Out · fafo.fm

Steve is such a great listen, the neurospicy 🌶️ rambles this episode goes on is so relatable. I feel like I really missed out on some great takes on intellij vs neovim, but got some really great knowledge about vector db’s, embedding, text compression, similarities to vector algegra like infinite craft.

Just popped open infinitecraft and I’ve definitely played this with my kids before, super fun, just could not remember the name of this one. I do remember an android one as well that is alchemist or something like that, which we have also played a lot.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 A quote from Mark Zuckerberg

2025-05-01T12:50:52Z

A quote from Mark Zuckerberg

You also mentioned the whole Chatbot Arena thing, which I think is interesting and points to the challenge around how you do benchmarking. How do you know what models are …

Interesting how confidently he says we can easily go to the top. really makes you wonder what we the normies are leaving on the table by using these general purpose models and what could be achieved with really tuned in models. Could I make an automatic blog tagger more accurately, maybe smaller, maybe tuned so well it runs fine on cpu?

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 r.jina.ai

2025-03-13T17:12:44Z

OpenTools | The API for LLM tool use

r.jina.ai

jina reader is a pretty sweet tool to convert a site to ai compatible text. There are other web to markdown types of tools, but the convenience of just adding r.jina.ai to the front of any page makes it so easy to grab for one page of docs.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Why did Framework build a desktop? - YouTube

2025-03-12T13:01:44Z

the racked up 4 framework mainboards sound wild. connected with usb4 and 5gig ethernet. they said they can run big models quantized down from 600Gb to within the 512GB limit they have. This seems wild to bring this level of capability to such a low price point. It will be really cool to start to see demos come out.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 OpenTools | The open MCP server registry

2025-02-23T22:31:33Z

One API to use any LLM with every MCP tool

OpenTools · opentools.com

OpenTools is an index for mcp servers to work with new agentic workflows like roo code and windsurf.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Proompted Kiddies Learning The Hard Way - YouTube

2025-02-14T19:15:41Z

Damn prime makes an interesting point near then end of this video. He’s seen a bunch of people able to just throw down charts and shit at their company and end up being “the coding guy” cause they proompted something once. In a way I can relate, I got into software in a similar way, but at a time that it took a lot more hard work, understanding , and copy past from the right stack overflow. Based on some of the people around me at the time I can only imagine how some people must feel like they got pushed into it without wanting it, and now are building something they don’t know anything about with no care about it or care to build any expertise. Is the future proompted charts from enterprise chatgpt or do we only continue growing more need for software from here.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 I Think I Love Deepseek R1 - YouTube

2025-02-01T16:29:43Z

There is a glimmer of hope out there that normal people can scrap together enough gpu to really run the latest models themselves. The ui really appears to be having huge leaps forward such that doing things like rag is no longer such a research project that it was just a few years ago. So excited to see Prime go through this homelab exercise.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 OpenAI Is Actually Terrible - YouTube

2025-01-31T03:46:29Z

Damn these deepseek memes go hard. Wild to see openai get played by their own game.

It’s crazy that the normie news that I have seen on deepseek shows that the Chinese made what the Americans did at a fraction of the price, without taking notice that they are building on the shoulders of openai.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 AI Is Making You An Illiterate Programmer - YouTube

2025-01-30T14:23:10Z

👏👏👏 This one is really good. I’m right there with him on most of this. I am very hesitant on subscription models, and all the ai tools feel like they are getting ready to be the next round of death by a thousand cuts, this time with pretty limited free tier and relatively high prices to run. I’m sure we will see companies get taken by huge bills soon by building off of someone else’s service.

On the flip side I’m definitely the guy that gets in a rut of just copy paste to the ai, wait for codeium to to inject. I feel like I have issues of momentum more than anything. When I’m on one side or the other I tend to stick it out for too long, but less so on going without because that llm drug is calling you when you hit a hard problem.

I’m excited to see him build out a homelab for llm stuff that he mentioned at the top. I’m interested, but probably not building one out for myself until we start to see some cheaper maybe used hardware to do it.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 The Worst Thing You Can Do For Your Career - YouTube

2025-01-17T14:54:14Z

Don’t stop learning! Stop trying because you have a doomer outlook on ai, llms, industry and think they are taking over. If you have no hope for the future, if you stop now you are cementing in that you will be no good and the ai will be better. Many, maybe most of us in this industry go here by hard work, long nights of learning, trying to solve problems that our job had. If llms take over then the world is going to be a whole lot different, it will be a world you cannot predict or plan for. For now put your head down and succeed in the world we have today.

TEEJ has some great thoughts on this whole sentiment, put this on for you morning walk or whatever you do.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Quick Start - kotaemon Docs

2025-01-11T16:34:26Z

Quick Start - kotaemon Docs

cinnamon.github.io

interesting UI for RAG based workflows, i.e. chatting with your documents. It looks like it can run a number of models, feels like ollama with RAG and a nice web ui.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 awesome-cursorrules/rules/py-fast-api/.cursorrules at main · P...

2025-01-09T14:03:12Z

Looks like a great start to a rules file for fastapi.

Place the happy path last in the function for improved readability.

I have never heard anyone say this. It feels weird to me. The other early return, find and handle errors early all make sense to me, but happy path last is new to me.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Generate Custom .cursorrules for Your Project Based on Communi...

2025-01-09T13:53:40Z

Generate Custom .cursorrules for Your Project Based on Community Examples

expert led courses for front-end web developers and teams that want to level up through straightforward and concise lessons on the most useful tools available.

egghead · egghead.io

Really interesting way to generate a rules file for agentic workflows based on your current repo. John uses gitingest here, looks like a fantastic tool, but probably not useful for most private repos. I’m sure you can replicate the same thing in a private repo wtih a small amount of effort the few times you need to do it. gitingest looks like a great way to pull in some extra context for some open source dependencies that you have though.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 cyclotruc/gitingest: Replace 'hub' with 'ingest' in any github...

2025-01-09T13:46:33Z

GitHub - coderamp-labs/gitingest: Replace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extract of a codebase

Replace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extract of a codebase - coderamp-labs/gitingest

GitHub · github.com

Gitingest has a python package on pypi that you can run with uvx, and it accepts the same arguments as the web version, right in your terminal

⬢ [devtainer] ❯ uvx gitingest --help
Usage: gitingest [OPTIONS] SOURCE

  Analyze a directory and create a text dump of its contents.

Options:
  -o, --output TEXT           Output file path (default: <repo_name>.txt in
                              current directory)
  -s, --max-size INTEGER      Maximum file size to process in bytes
  -e, --exclude-pattern TEXT  Patterns to exclude
  -i, --include-pattern TEXT  Patterns to include
  --help                      Show this message and exit.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Gitingest

2025-01-09T13:40:41Z

Gitingest

Replace

gitingest.com

Replace hub with ingest in any github and get a prompt friendly codebase ready to feed into any llm. It combines the entire codebase, based on a gitignore style glob that you pass in, into a single TXT file.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Replicate — Run AI with an API

2024-06-01T00:59:49Z

Replicate - Run AI with an API

Run open-source machine learning models with a cloud API

replicate.com

This is so easy compared to self hosting stable diffusion yourself. It even has a nice api that you can hit with curl or python. The pricing seems competitive as well. Bookmarking this to try next time I need something like it.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

How to kill ollama server

2024-04-14T18:31:36Z

I recently updated ollama, and it now installs a systemd service that I was not expecting. Seems like a great option, but I hadn’t expeted this and I was able to kill it previously. It was using up gpu, and I do other things on my machine with a gpu. I tried pkill, kill, and everything, it was still coming back.

No matter what it comes back

# stop it
systemctl stop ollama.service

# disable it if you want
systemctl disable ollama.service

# confirm its status
systemctl status ollama.service

You can confirm this with the following command.

# checking running processes
ps aux | grep ollama
pgrep ollama

# checking gpu processes
gpustat --show-cmd --show-pid

Next time you want to start you can do it as before with ollama serve.

💭 Use an llm to automagically generate meaningful git commit mes...

2024-04-11T13:35:50Z

Use an llm to automagically generate meaningful git commit messages

harper.blog

This is pretty sick, I wanted this early on when I was making lockhart. I wanted to do the git hook thing but could not figure it out and did not know that prepare-commit-msg was a hook that I could use.

Git Hooked Then I remembered! Git hooks! Lol. Why would I have that in my brain - who knows!

I asked claude again, and they whipped up a simple script that would act as a hook that triggers with the prepare-commit-msg event.

This is awesome, cuz if you want to add a git message, you can skip the hook. But if you are lazy, you exclude the message and it will call the LLM.

Simon Willison’s llm cli comes in clutch here, it has such a good intereface to allow a prompt to be piped in, but the system prompt be set by -s.

gpt = "!f() { git diff $1 | llm -s \"$(cat ~/.config/prompts/commit-system-prompt.txt)\" }; f"

I love hacking on projects, but often I am super bad at making commits that make sense.

I completely relate to this statement, and this is why I am trying it.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Replace Github Copilot with a Local LLM - YouTube

2024-01-30T23:14:01Z

This really makes me want to try Dolphin Mixtral with ollama now. It looks very impressive from this video. The ability to keep adding features before becoming confused is though with a lot of these llms.

Being chat based, this is not a co pilot replacement. I was really hoping for an in line co pilot like tool that I can run locally. I have not used co pilot yet, but I have had great luck with codeium.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 charmbracelet/mods: AI on the command line

2024-01-26T02:15:42Z

GitHub - charmbracelet/mods: AI on the command line

AI on the command line. Contribute to charmbracelet/mods development by creating an account on GitHub.

GitHub · github.com

This is a pretty sweet interface into llms. I used it a bit with my son tonight while he was asking me for datapack ideas.

❯ mods -f 'I am trying to have fun on my minecraft server and am creating a minecraft datapack send me some load.mcfuncions that will make it fun'

You can continue the conversation with a -C

❯ mods -C -f 'I like where you are going with number 4, can you make it so that it runs when a player opens a door'

You can pass it some data

curl https://waylonwalker.com/thoughts-on-unit-tests/ | mods -f 'summarize this post'

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 David-Kunz/gen.nvim: Neovim plugin to generate text using LLMs...

2023-10-14T02:31:29Z

This seems like a promising tool to use with ollama.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Ollama

2023-10-14T02:31:03Z

Ollama

Ollama is the easiest way to automate your work using open models, while keeping your data safe.

ollama.ai

ollama is the easiest to get going local llm tool that I have tried, and seems to be crazy fast. It feels faster than chat gpt, which has not been the experience I have had previously with running llm’s on my hardware.

curl https://i.jpillora.com/jmorganca/ollama | bash
ollama serve
ollama run mistral
ollama run codellama:7b-code
ollama list

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 Simon Willison on X: "Anyone got a lead on a good embedding mo...

2023-09-10T00:05:15Z

X (formerly Twitter) · twitter.com

Kinda mindblown that this is even possible. This is so far outside of my current thinking that i didn’t even think of an elegant way to implement semantic search accross images and text at the same time. I know it happens at Google, but I envision that as still text search accross tags and meta data about the image.

Based on the number of responses CLIP is the thing that does this.

Note

This post is a thought. It’s a short note that I make about someone else’s content online #thoughts

💭 LLM now provides tools for working with embeddings

2023-09-07T00:35:55Z

LLM now provides tools for working with embeddings

LLM is my Python library and command-line tool for working with language models. I just released LLM 0.9 with a new set of features that extend LLM to provide tools …