Chris Raethke (@codesoda) header
Chris Raethke (@codesoda) profile picture

Chris Raethke

@codesoda

Improving communication using product thinking and machine learning. Prev @NotivHQ @bugcrowd. Purveyor of bad dad jokes! #peoplefirst

 San Francisco, CA
codesoda.com
Joined December 2008
  • Tweets4.7K
  • Following1K
  • Followers2.6K
136Media
Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)Photo by Chris Raethke (@codesoda)
I love that @antiburn_ai helps me cut so close to my usage limits. Have now swapped fable 5.1 medium for opus 5 high until the morning. #Tokenomics
Image from Chris Raethke (@codesoda)
1
3
291
Anyone else have a mispent weekend burning tokens? github.com/codesoda/openjev-… #jev
GitHub - codesoda/openjev-rs: Local typed decisions from frozen GGUF model logits: Rust workspace...

GitHub - codesoda/openjev-rs: Local typed decisions from frozen GGUF model logits: Rust workspace...

Local typed decisions from frozen GGUF model logits: Rust workspace and CLI - codesoda/openjev-rs

github.com
1
6
335
It hosts a jev compatible http api. Tonight I'm actually going to rename this to system1-rs and update it so you can change the backend, so you can swap between GLiNER2.5, SemLf, Laya or proxy to Jev via vercel/openrouter.
1
3
128
It's happening github.com/codesoda/systemon…
GitHub - codesoda/systemone: One CLI and Jev-compatible API for local and hosted typed-decision...

GitHub - codesoda/systemone: One CLI and Jev-compatible API for local and hosted typed-decision...

One CLI and Jev-compatible API for local and hosted typed-decision backends (planning) - codesoda/systemone

github.com
1
36
Every person / team should own their harness, whether that’s optimizing a vendor one or rolling their own. The better you understand the internals, the better you’ll be at driving it.
geoff (@GeoffreyHuntley) profile picturegeoff
@GeoffreyHuntleySep 10
I dream for a future in our industry where good/great engineers, rock into a hiring pipeline, and when they respond to the “what AI tool they use” question, the response is…. “I use my own. Wanna see it? Let’s nerd!”
2
361
Anyone using @grok @build that is keen to test antiburn.ai for me?
antiburn — Stop hitting your token limits

antiburn — Stop hitting your token limits

Where your tokens went, and how to stop burning through your quota. Straight from your session logs.

antiburn.com
3
1
124
Everyone is freaking out about @claudeai Fable using up their allowances. You're doing it wrong!
Chris Raethke (@codesoda) profile pictureChris Raethke@codesodaJun 10
Replying to @Shpigford
The solution is you don't use fable for everything, set a haiku co-ordinator, sonnet/opus for planning/review and fable ONLY for the implementation. The sonnet/opus can review it, then get fable to update implementation if required. Haiku then pushes and creates PR, fixes CI 🏁
1
89
Dropped a project today: Discuss CLI No more reviewing agent plans in the terminal. Discuss CLI turns any Markdown file into a browser review surface with PR-style comments — and your coding agent can join, read the exact text, and reply in the margins. github.com/codesoda/discuss-…
2
1
3
232
I kept manually asking my coding agent to verify database state and click through flows after every change. So I built Bugatti.dev. "Submit the signup form and verify the user appears in the database" — that's your test. Your agent runs it, you get pass/fail.

Write tests in plain English. An AI agent runs them.

A CLI test harness that drives AI agents through plain-English test plans. Structured pass/fail results, built-in infrastructure, full audit trail.

bugatti.dev
Chris Raethke (@codesoda) profile pictureChris Raethke@codesodaMar 27
Building a new thing tonight, would love your feedback. Plain-English local QA, powered by your coding agent. Describe the flow in TOML, Bugatti clicks the buttons. github.com/codesoda/bugatti-…
docs: add product requirements document by codesoda · Pull Request #1 · codesoda/bugatti-cli

docs: add product requirements document by codesoda · Pull Request #1 · codesoda/bugatti-cli

Summary Adds the product requirements document (PRD) for bugatti-cli, outlining goals, features, architecture, and milestones Test plan Review PRD content for completeness and accuracy

github.com
1
1
157
Image from Chris Raethke (@codesoda)
64
Building a new thing tonight, would love your feedback. Plain-English local QA, powered by your coding agent. Describe the flow in TOML, Bugatti clicks the buttons. github.com/codesoda/bugatti-…
docs: add product requirements document by codesoda · Pull Request #1 · codesoda/bugatti-cli

docs: add product requirements document by codesoda · Pull Request #1 · codesoda/bugatti-cli

Summary Adds the product requirements document (PRD) for bugatti-cli, outlining goals, features, architecture, and milestones Test plan Review PRD content for completeness and accuracy

github.com
1
1
2
251
Your coding agent accumulates patterns you never notice. Skillable mines your session transcripts to find repeating workflows and turn them into skills. Works with Claude Code, Codex, OpenCode + 9 more. github.com/codesoda/skillabl…
GitHub - codesoda/skillable: Claude Code skill — Analyze tool call patterns across sessions to...

GitHub - codesoda/skillable: Claude Code skill — Analyze tool call patterns across sessions to...

Claude Code skill — Analyze tool call patterns across sessions to discover common workflows and suggest new custom skills - codesoda/skillable

github.com
2
127
Currently ralping a new thing AI first bookmarking - What's interesting is not so much in what it is, but how it's coming into fruition. github.com/codesoda/agentmar…
GitHub - codesoda/agentmark

GitHub - codesoda/agentmark

Contribute to codesoda/agentmark development by creating an account on GitHub.

github.com
1
2
78
I worked with GPT 5.4 to brainstorm, then codex and I wrote a PRD (see tasks). That was then strategically broken up into detailed specs using my auto-ralph skill, which I reviewed meticulously. Now auto-ralph is [planning, buidling, reviewing]+ through each spec to build it out.
1
1
94
I love that I can just follow along as it builds it out. github.com/codesoda/agentmar…
feat: Ralphy Mc Ralph Ralph (building it) by codesoda · Pull Request #1 · codesoda/agentmark

feat: Ralphy Mc Ralph Ralph (building it) by codesoda · Pull Request #1 · codesoda/agentmark

Iteration 93: Spec 25 - Install Script Summary Implemented a complete POSIX shell installer (install.sh) and comprehensive test suite (install_script_tests.rs) for spec 25. The installer builds and...

github.com
42
You know what would be nice, different permission profiles inside a @claudeai project, e.g. I'm in dev profile, so that has a bunch of skills, tools, etc., enabled. I'm in the AWS/DevOps profile, so I enable all the default things for that. But in general, it's locked down.
87
Couldn’t have said this better myself!
Mat Sherman (@Mat_Sherman) profile pictureMat Sherman
@Mat_Sherman1 Jun 2025
You don’t have to raise money in SF, but the investors you seek to work with should have made their money from that ecosystem. Raising is rarely about one round. The network needs to run very deep to keep the flame lit. One round is rarely enough.
210
Ever noticed how funny it can be to build a voice app in an office/coworking environment? People think you’re randomly talking to yourself all day!
2
137
Chris Raethke retweeted
Hiten Shah (@hnshah) profile picture
Hiten Shah
@hnshah16 Feb 2025
Your product is only as good as how fast a user can say, "Oh, I get it."
14
21
3
184
14,875
Chris Raethke retweeted
mr fabio (@fffabs) profile picture
mr fabio
@fffabs5 Feb 2025
My daily reminder
Image from mr fabio (@fffabs)
351
9,604
1,375
69,699
10,820,458
Load more