AIchemist

Mixing models, agents, and workflows until something useful appears. Watching AI go from magic trick to everyday leverage 🧠

[AIchemist @MrSplitSignal]

@gdb Not gonna lie, I’m genuinely sad to lose Atlas Browser. Chrome with the sidebar just isn’t the same, and the built-in browser in Codex doesn’t quite feel like a true browser experience yet. Really hoping that part of the product continues to evolve as everything matures.

[AIchemist @MrSplitSignal]

@sama A new kind of computer you say? 👀

[AIchemist @MrSplitSignal]

@nickfrosst @cohere Great to see you supporting this Nick!

[Matt Shumer @mattshumer_]

Claude Opus 5 one-shotted this game.

EVERYTHING you see in this demo is custom code... not a single external asset was used.

AI games are going to be amazing.

(sound on)

[AIchemist @MrSplitSignal]

@ajambrosino Merge them into a single, unified experience.

[AIchemist @MrSplitSignal]

@_simonsmith Would be cool to see its chain of thought appear in real time as you’re talking with it. @athyuttamre

[AIchemist @MrSplitSignal]

@JustinBleuel @athyuttamre @athyuttamre @JusinBleuel Appreciate you looking into this.

[AIchemist @MrSplitSignal]

I’m curious about when OpenAI might introduce live video sharing with the updated voice model.
@JustinBleuel

[AIchemist @MrSplitSignal]

@guinnesschen This is seriously impressive. Can’t wait for it to hit my account.

[M1 @M1Astra]

Early preparations for Claude Opus 5 have been spotted again. Seems imminent.

[AIchemist @MrSplitSignal]

@JustinBleuel @ChatGPTapp Love this. This was one of my favorite features of Co-work. Glad you guys brought it to ChatGPT!

[AIchemist @MrSplitSignal]

- Bring meeting recorder to the iOS app.
- Use a single interface rather than separate Chat and Work.
- Add a Home interface to control Hue lights and Mission Control for the home.
- Bring back Pulse; it was my favorite Pro feature.
- Make the new bidirectional voice update work with live video.

[ClaudeDevs @ClaudeDevs]

We're also keeping Claude Code weekly limits 50% higher, now through August 19, for all Pro, Max, Team, and seat-based Enterprise users.

[AIchemist @MrSplitSignal]

@Dimillian Totally fine when you've got remote Codex on mobile 🫡

[AIchemist @MrSplitSignal]

@DanJCleary @AnthropicAI Congrats Dan, crush it!

[AIchemist @MrSplitSignal]

@tarstarr @aaronp613 @thsottiaux makes sense Tara. Appreciate the quick reply. I think being intentional about it will make the experience a lot stickier for ppl. My biggest ask would be bringing this into the iOS app. I can already think of a ton of use cases.

[AIchemist @MrSplitSignal]

@_catwu Would be great to do a Claude Code session and show your team my workflow, but only if/when those sessions become available. Thanks!

[leo 🐾 @synthwavedd]

🚨 BREAKING: Kimi K3 benchmarks have released and it's ~Fable/Sol level

[Boris Cherny @bcherny]

Something I have been thinking about: in the past, the best engineers I knew spent a lot of time automating their work in various ways. Better vim/emacs automations, writing lint rules to catch repeat code issues, building up a suite of e2e tests so they don't need to smoke test the app manually. These kinds of things were the highest leverage activities an engineer could do, because it multiplied their own output, which in turn meant they could build more things.

I think many of these automations have become even more important now. This is true for a number of reasons.

First, infra and DevX automation speeds you up. And if you are running an army of agents, each of those agents will be sped up also. More automation == more output per unit of time.

Second, moving things to code improves efficiency. Your agent could fix an issue every time it sees that issue happen, but that uses tokens and might miss cases. If Claude instead writes a lint rule, CI step, or routine, that class of issue can be fully automated forever. This is really what people are talking about when they talk about loops -- it's about automating entire types of busywork rather than solving them one off. This isn't a new idea at all. Engineers have been doing this for a long time!

Third and most importantly, automation makes it possible for others to contribute to the codebase more easily. Increasingly what I am seeing is engineers are contributing to codebases on day one because Claude can navigate the codebase for them, and that non-engineers are able to contribute to a codebase as effectively as engineers can. What gets in the way of both of these is domain knowledge that lives in peoples' heads rather than in automation -- the stuff you used to have to learn when ramping up. What has changed thanks to agents is the domain knowledge that can be encoded as infrastructure is no longer limited to what is expressible in lint rules and types and tests; it can now capture nearly all domain knowledge, encoded as code comments and skills and CLAUDE.md rules and memories. If I put up a PR for an iOS codebase I don't know and a code reviewer rejects it because it doesn't use the right framework, or if a designer builds a new feature and it gets rejected because it doesn't follow the right architectural patterns, these are failures of automation.

Every team should be writing the CLAUDE.md's, REVIEW.md's, skills, and docs that enable agents to productively work in their codebase with zero additional context from the prompter. This sounds crazy, and at the same time is a natural extension of the stuff engineers have always done: automate, and encode domain knowledge as infrastructure. As the model gets smarter and as the harness matures, this task becomes easier. In the meantime, it is on every team to look for ways to convert their domain knowledge to infra so that Claude can write code better, so that code review catches issues automatically, and so the next person working on your codebase can contribute more easily.