Zach Cox @zcox

🏔️🤖 theza.ch Draper, UT Joined January 2008

Tweets

4K
Followers

743
Following

1K
Likes

436

Zach Cox @zcox

a week ago

@runarorama underrated. claude code has made me a git power-user. i personally despise the git cli. for anything more than add/commit/push i now use an agent, and can accomplish so much more than i ever could before.

0 0 1 44 0

View Details

Zach Cox @zcox

a month ago

opus for show, sonnet for a pro

0 0 0 59 0

View Details

Zach Cox @zcox

3 months ago

i'll buy an @AnthropicAI eng their fav drink to RL "add target/ to .gitignore in new Rust project before first git commit" into claude

0 0 0 128 0

View Details

Zach Cox @zcox

3 months ago

@TheVixhal progressive disclosure ftw! very cool

0 0 0 16 0

View Details

Zach Cox @zcox

4 months ago

The last GA (non-preview) release of an LLM from Google was gemini-2.5 in June 2025. 9 months ago. Since then we've seen Claude 4.5/4.6 and GPT 5/5.1/5.2/5.3/5.4. Why isn't Google releasing any more models in GA?

0 0 0 156 0

View Details

Zach Cox @zcox

4 months ago

coding agents won't kill open source which is faster? 1. fully research, spec, plan, generate, and verify your own bespoke standards-compliant http stack and web application framework in Rust 2. use axum repeat for json, sql, s3, kafka, OS, containers, k8s, etc

0 0 1 117 0

View Details

Zach Cox @zcox

4 months ago

0 0 0 114 0

View Details

Zach Cox @zcox

4 months ago

I prefer "verification" over "backpressure" for coding agents backpressure implies resistance. but we want faster loops, not slower verification provides signal back to the generate step. fewer, faster iterations to convergence

0 0 0 95 0

View Details

Zach Cox @zcox

5 months ago

example: github.com/zcox/rust-span…

0 0 0 48 0

View Details

Zach Cox @zcox

5 months ago

dir structure looks like:

1 0 0 58 0

View Details

Zach Cox @zcox

5 months ago

a single IMPLEMENTATION_PLAN .md grows unbounded and wastes context, or requires constant lossy maintenance instead, decompose tasks within specs so agents can explore via progressive disclosure and track status via emojis: 🔲 Not started 🔄 In progress ✅ Complete

1 0 0 84 0

View Details

Zach Cox @zcox

5 months ago

@headinthebox while you are still experimenting with a loop I extended mine to a spiral and added rotation in a second orthogonal plane to form a clifford torus. my claude Code now optimizes skills across the multiverse.

0 0 2 132 0

View Details

Zach Cox @zcox

5 months ago

OG surely has better magic spells than me but I've made it pretty far reading posts, watching videos, putting in reps, incrementing weight, burning tokens the techniques are effective on 5-10yr old brownfield enterprise Scala services. but always hungry to learn more!

geoff @ 4th of july americana @GeoffreyHuntley

5 months ago

as soon as i post it corps will rebundle it and sell it where’s the upside for us et al topics heck i even had a frontier lab in my DMs after posting the above tweet i was seriously considering moving to SF to join them but they said no but now they are in my DMs asking

8 1 26 8K 7

0 0 0 97 0

View Details

Zach Cox @zcox

5 months ago

And then CC responds to that with a super dope haiku about penguins.

0 0 0 46 0

View Details

Zach Cox @zcox

5 months ago

You can see that CC called the Skill tool with argument "skill: write-dope-haiku" and the result is only "Launching skill: write-dope-haiku". The following user message (which I didn't send) contains: Base directory for this skill: /path/to/skill <contents of SKILL .md file>

1 0 0 55 0

View Details

Zach Cox @zcox

5 months ago

I assumed Claude Code put the skill content in a tool result, but it's actually in a synthetic user message sent after a tool call. Using @LangChain LangSmith's Claude Code hook you can see the tool call & user message. I'm curious why the user message? Drives the model forward more effectively than a tool result?

1 0 0 139 0

View Details

Zach Cox @zcox

5 months ago

1. Ralph correctly 2. optimize loop fewer faster iterations common slow parts: - finding next task - starting/stopping servers, dbs, etc - git commit after task completion - Claude's affinity for giant summarization of accomplishments ruthlessly minimize or convert to scripts

0 1 1 109 0

View Details

Zach Cox @zcox

5 months ago

I really like the progressive disclosure structure of Loom's specs github.com/ghuntley/loom/… Seems like a good idea for plans/tasks too. Single IMPLEMENTATION_PLAN .md gets too big.

0 0 0 75 0

View Details

Zach Cox @zcox

5 months ago

agent memory in action

0 0 0 38 0

View Details

Zach Cox @zcox

5 months ago

key component of GH's Ralph: self-updating status checklist plan .md contains: # Status - [x] Phase 1 - [ ] Phase 2 - [ ] Phase 3 prompt .md contains: study @ plan .md complete the next task only then update the status checklist cat prompt .md | claude