Waylon Walker

Agents Are Here - Scaffolding Document

Waylon Walker avatar
Author
5 min read
Copy this post

Agents Are Here - Scaffolding Document

Use this as a blueprint for completing your post. Cross-links marked with [[…]], suggested insertions marked with TODO.

Current Draft Structure (with additions) #

Opening / Late 2025 #

Current: References I'm Out On Agents

TODO: Add Sep-Nov context

Before the breakthrough, there was [[when-to-vibe-code]]—back when "vibe coding" 
meant accepting code you didn't read. Theo's framework mattered then: don't 
land in "don't know, don't care, mission critical." Most of us were stuck there.

Link opportunity: Developer Vs Artist Ai - “Developers got great tools… artists were given tools for developers to do art with.”

The Inflection Point / December 2025 #

Current: Mentions Anthropic Opus 4.5, December slowdown

TODO: Add tooling explosion

December wasn't just about the models getting better. The tooling exploded.
I started noticing [[opencode-variants]] popping up—fast, slow, thinking modes. 
The free Zen provider made it accessible in ways it hadn't been.

Link opportunity: I'm In-ish On Agents - “Context is king, good plans are paramount, syntax barely matter.”

January 2026 #

Current: “flu season”, wiped out but wanting to work

TODO: Add the mindset shift

January was when I wrote [[stop-using-boomer-ai]]. The chat-copy-paste era 
ended for me. If you were still doing that, you were doing it wrong.

Somewhere in the fever haze, I started figuring out the harness. Not just 
prompting—planning. [[my-first-agentic-workflow]] documents the /init, AGENTS.md, 
the whole ritual: issue → plan → execute → review.

Quote to weave in from I'm In-ish On Agents:

“Around Aug 2025 I wrote I'm Out On Agents… This was just after the tipping point where not just hype bros were making the switch, but people I work with, people I trust… People like the creator of Flask Armin Ronacher.”

February 2026 #

Current: “what just happened?” (empty)

TODO: This is your biggest gap. Content from Agent Management Is Exhausting:

## February 2026
_what just happened?_

It got fast. Too fast.

[[agent-management-is-exhausting]]

Claude could implement features faster than I could research and raise issues. 
It's like trying to speedrun a Minecraft seed when you just figured out how 
to craft a pickaxe.

The exhaustion was real—managing these things stretches a different part of 
your brain than you're used to using.

It also became clear that [[pm-not-babysitter|babysitting was the wrong frame]]. 
Theo's quote haunted me: 10 lines of code per day for humans, 10k LOC for agents. 
Stop being the babysitter, start being the PM.

Link opportunity: How To Run 5 Agents In Parallel Feb 2026 Edition - “Yes, developers are running 5 agents in parallel… It requires planning, it requires tooling.”

yes or –dangerously-accept #

Current: Empty section

TODO: Bridge February learning → March reality

## yes or --dangerously-accept

Somewhere between February's chaos and March's clarity, the workflow solidified. 
The yes-or-die moment. The --dangerously-accept flag.

This is when you decide: are you reviewing every line, or are you managing plans?

From [[pm-not-babysitter]]: "You need a tool for planning and tracking, otherwise 
you are playing babysitter rather than Product Manager."

March 2026 #

Current: Empty section

TODO: Add productivity tension from [[thinking-about-ai-productivity-again]]:

## March 2026
_the productivity paradox_

I built one of the biggest PRs I've ever done professionally. Fifty commands 
refactored, stdout/stderr contracts established, Unix-pipe friendly. 
All patterns from clig.dev, implemented consistently.

The agents cranked out more code than I could have typed in months.

But here's the thing: [[thinking-about-ai-productivity-again|it's low value work]]. 
We wouldn't have put humans on this wholesale. Not cutting costs, not selling 
more product. Just... making things consistent and boring.

Massive amounts of work got done. But was it the *right* work?

Also from I don't want someone else running my agents:

Around this time, I realized: I don't want someone else running my agents. 
I don't want to review the mass of changes clobbered across the codebase. 

"If someone is going to be stirring the slop in my product, I want it to be me."

April 2026 #

Current: Empty section

TODO: Current state / where we are now

## April 2026
_here now_

The agents are here. They're not what we thought they'd be in 2024, not what 
I thought they'd be in August 2025. 

They're exhausting. They're fast. They require a different kind of management.

From [[is-ai-faster-yet]]: "I'm definitely doing more... but hard to sus out 
real productivity from the noise."

But they're here. And we're still figuring out what this job looks like.

Missing Month: September-November 2025 #

Suggestion: Consider a short section before “The Inflection Point” covering the “vibe code” era. Context from When To Vibe 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.

This sets up the contrast with where you are now.

Quote Bank (for weaving throughout) #

From I'm In-ish On Agents:

  • “Context is king, good plans are paramount, syntax barely matter.”
  • “There is no free lunch. Software engineering is still very much needed, but the work is switching.”

From Agent Management Is Exhausting:

  • “The secret I’ve discovered is a good, well-defined plan.”
  • “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.”
  • “I had a session yesterday where the context got poisoned with a wrong assumption… 20 minutes of perfectly executed code solving the wrong problem.”

From Pm Not Babysitter:

  • “Stop babysitting your agents, treat them like a real team and they will reward you.”
  • “Your agents are not doing 10k lines with you babysitting them.”

From How To Run 5 Agents In Parallel Feb 2026 Edition:

  • “Planning is the core of what it takes to keep agents running… Agents need something to do, telling them to turn the circle green, then blue, then to a rectangle, is not it.”
  • “Is anyone doing this all day? Probably not.”

From Is Ai Faster Yet:

  • “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.”
  • “Everything feels so new and different that there’s a bit of a learning curve to understand it all.”

From I don't want someone else running my agents:

  • “Work is feeling different now. New lines need to be drawn in new directions.”
  • “Expectations are changing, the way work is completed is changing.”

From [[thinking-about-ai-productivity-again]]:

  • “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.”

Own transition posts:

Older context:

TIL/tooling:

Next Steps #

  1. Fill in February (biggest gap currently)
  2. Add the March productivity paradox section
  3. Complete April - current state reflection
  4. Optionally add Sep-Nov 2025 “vibe code era” context
  5. Ensure all TODO blocks are either used or removed
  6. Run markata-go build to verify all cross-links resolve

Connections

Related tags and posts connected to this entry.