w h i t (@whit) header
w h i t (@whit) profile picture

(he/him) skeets at whitmo . b s k y .social x erased my alt-werk xeets account... System maker and breaker, hire me!

Bend, OR
Joined December 2008
Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)Photo by w h i t (@whit)
Listen to this. And then listen again. And again. Until you get it.
1,162
9,826
506
34,737
1,986,842
I just released Patchmill v1.0 It is a complete rebuild. The old workflow engine is gone, and a small kernel takes its place: it starts Pi, claims issues, reads git and the issue host, and checks what the agent actually did. Everything that needs judgment, such as how to plan, code, review, and write, now lives in skills you can read and change. The production code went from 53,542 lines to 4,300. On top of that kernel, this release adds two things we've wanted for a long time. You can now plan an issue with Patchmill in a chat, approve the spec from a dialog, and let a background worker build it. And you can run Patchmill without touching your tracker's labels at all. The skill pack also grew: frontend design with real HTML mockups, an independent frontend review, plain-English writing rules, and an implementation skill that always gets a fresh reviewer. This is a breaking release. There is no upgrade path from 0.20 configuration or saved state. Read Upgrading below before you run it on an existing repository. github.com/rochecompaan/patc…
1
2
69
w h i t retweeted
Is this management? Am I a manager now??
148
135
26
2,981
66,689
the most import skill in 2026 is still being able to tell your agents "wtf is this crap, no, not like this" if you stop doing that, you're basically fucked, despite what some devfluencers would like you to think.
209
415
84
4,209
209,214
Warren 2028
55
curious how this is effected by N humans. hypothesis: for N=0-1 Agent adversarial review mostly effective N=2 Spy vs Spy, but still pretty effective N>2 Spy^N creates toxic context and time fragmentation Quality plummets.
I had my agents analyze thousands of recent adversarial reviews in my system. Something like a quarter overall were blocked by the reviewer, and a meta-review showed the issues raised by agentic reviewers were almost all legit. I disallow filing P3s, which helps a lot. But in a nutshell, they work for me and I'm keeping them.
72
w h i t retweeted
Asked one of our developers how he uses his AI agents and he just sent me this:
179
1,337
270
16,199
1,778,162
the saddest part of the headless internet is that junior devs will never know the beautiful and elegant intricacies of the AWS console
32
16
4
563
17,459
AI agents, because I need more embarrassment in my life.
39
Fear of 0day is like being terrified of ninjas instead of cardiovascular disease.
23
441
33
856
I want this on a t-shirt
Complexity is a ratchet.
1
72
wondering if the main danger of intelligence comes from the tendency to overestimate ones capabilities.
94
me when things works
This banger sums up Ohio
1
115
lolgag
Rust is the Perl of the 21st century.
41
resonance
Some mistakes you might be making: Your effort level is too high Your skills are too long You're not aligning before you start work You're aligning too often, and too hard Your automated checks are too lax Your custom tooling is too token-inefficient You aren't doing enough automated review You aren't reviewing previous coding sessions (with an agent) You're doing too much human review You aren't vibe coding enough prototypes Your agent can't verify its own work You're babysitting the agent too much You've overloaded your global steering files Your codebase is hard to navigate You aren't building a shared language Your codebase architecture allows bad tests to be written You aren't thinking strategically enough You aren't thinking enough
41
enter the AIREs or if you of that ilk, SIREs.
we'll get AI SREs that are superhuman by today's standards, but we'll still create problems faster than we can solve them. what holmes gives, claude takes away.
41
Snow on Tumalo Mountain this morning
18
lowkey i wonder if our model could actually be like really useful for weighted monte carlo search in proofs, like; finding the most promising directions then escalating to reasoning model periodically
1
260
there is a particular kind of personality disorder than gravitates toward mistreat waitstaff.
1
80
.
128
1,309
126
16,123
475,713