A graph you can act on, not just stare at 🌴 @obsidian.md

Still great advice 43 years later

What if Anthropic is intentionally making Claude sound like the most annoying TED talk presenter ever so that we know it's not a real person 🤔

Got to talk at @aidotengineer.bsky.social conf last week about the need for collaborative AI engineering. All our current coding agents are single player. We're trying to scale up individual productivity, but creating tons of alignment problems in the process. We have no good tools for...

Every year, I think, “This must be the most interesting year to work on AI coding agents,” and every year, I am correct.

AI can create code fast. But it does not mean you cannot and should not understand it at the speed at which it creates it.

This is how I use AI to augment my design process. Instead of having the agent show me a giant diff of changes, Atelier links each edit back to the original thread of feedback so I can review them in context

There comes a point in every man’s life where he releases an open source frontend framework. That day, for me, is today. Tutuca: a zero-dependency, batteries-included SPA framework marianoguerra.github.io/tutuca/ Would love your feedback on the framework, the landing page and the tutorial

The CLI is great for a lot of things but reviewing Claude Code plans is not one of them. So much scrolling back and forward just to figure out what the agent changed

I just released v0.3 of atelier.dev with the new inline revisions feature available for everyone to try

Revising a document with an agent should feel like collaborating with someone sitting next to you reading the same page. You highlight what needs to change, the agent suggests revisions, and you decide. That's what I'm building in Atelier

Revising a document with an agent should feel like collaborating with someone sitting next to you reading the same page. You highlight what needs to change, the agent suggests revisions, and you decide. That's what I'm building in Atelier

Last night the Resonant Computing Lab launched in NYC. A fund for building software that leaves people nourished, not drained. 1,200+ people have signed the manifesto. Now there's funding to turn principles into practice. resonantcomputinglab.com

The gap between Hack and NetHack taught me something about the future of CS. Worth thinking about. Does Computer Science Still Exist? davidbau.com/archives/20...

drop + merge mini apps on canvas. v0

I want my workflow to feel more like this. One big reconfigurable space for deep work

When the agent finishes a task I launch atelier-verify to automatically verify the acceptance criteria and move it to the done column if they're all met

The deepen-plan skill from Compound Engineering is another nice pattern to add extra effort when you need it instead of having to define it up-front github.com/EveryInc/com...

Not a fan of this new effort selector in Claude Code. It's cool that the models can adapt to different tasks but it's annoying that you need to think about which setting to use before you send each prompt. Why can't I just define which tasks are high effort in CLAUDE.md?

I added realtime notification so that when I'm running multiple agents in parallel I can scan the board and see which ones need my attention. Clicking on the yellow text automatically switches to the terminal for that agent

There is a difference between how we approach work where we’re given known constraints, known capabilities, vs work where we get to explore the possibilities of the new tech & define the constraints & capabilities. The latter is science led, and the process is Applied Research.

The biggest change in Atelier.dev v0.2 is that spec files now open in a custom Markdown editor. So now every task gets its own "plan mode"

I've been thinking about why verifying AI agent output feels so much harder than writing the spec that produced it. That question led me to rethink where my attention actually belongs in the process, and eventually to build atelier.dev narphorium.com/blog/decisio...

I've been thinking about why verifying AI agent output feels so much harder than writing the spec that produced it. That question led me to rethink where my attention actually belongs in the process, and eventually to build atelier.dev narphorium.com/blog/decisio...

I wish they made this with an oled display on the outside and a foldable e-ink display on the inside

Softerware (new post) share.google/T1AajhhxRprb...

with all the deskilling and gell-manning (etc) we're gonna get from automated thinking machines, now is the time for the 'tools for thought' crowd to make tools that help you think things through for yourself. tools to do *more* hard thinking. it should feel effortful. cherish that feeling.

Values-as-Spec is the process of specifying what values the product is committed to protecting; how the product involves the user in its processes; and what the product explicitly will not automate or replace. What it does, won’t do, and can’t do. @seanx.bsky.social dearhermes.com/read/kfniw9y...

Self-modifying AI agents: "There are no configuration files to learn. Just tell Claude Code what you want. Or run /customize for guided changes. The codebase is small enough that Claude can safely modify it." github.com/qwibitai/nan...

If you're a library author, it's a pain to figure publish agent skills in a way that your users will actually see. If you're a developer using libs, it's a pain to make sure you're using the best and latest agent guidance for them. Let's solve it! I'm proposing .agents.lib

📝 My contribution to the discourse on the software industry's favorite topic du jour. seanvoisen.com/writing/prog...

Recently, I put all of my team's first-principles thinking about the future of computing into one demo. It's our vision of the capabilities a truly AI-native personal computer will have. ruperts.world/blog/ai-comp...

I find it funny when Claude tries to talk me out of building something like "you _could_ build it that way but it would probably take a month" and then builds it in a hour

Atelier started as a set of custom agent skills that I've been iterating on as I moved from chat-based development to spec-driven development. The kanban board is how I organize it all, but it's just a window onto the workflow.

No tool is magically going to save you from the technical debt of vibe coding. Verifiability has to be baked into every step. Atelier attaches specific acceptance criteria to every task and automatically verifies them.

One of my favorite little details is that you can hover over blocked tasks to quickly see the blocking tasks or hover over a plan to see which subtasks are completed

The new Agent Teams feature in Claude Code lets multiple agents work simultaneously. Fun watching the atelier.dev cards update automatically as each finishes: code.claude.com/docs/en/agen...

I added a direct link to the .vsix file for Cursor users until atelier.dev is available in the Cursor extension registry

Going fast with AI is less about generating more code. It’s more about gardening a working system whose complexity scales linearly as it grows.

I built Atelier because I needed a better way to organize specs for my agent workflows. When I click Implement on a task, it runs Claude Code with my custom prompt template. When Claude finishes, the card automatically moves to review. youtu.be/eVZJc7Rz8OI

I put a lot of work into this release 😊 excited to share the new features! www.spacecake.ai #ClaudeCode #opensource

Would love to see this get widely adopted. Having deep links between code and conversation history would be so valuable for both developers and agents agent-trace.dev

Under the hood, Atelier uses Claude Code's new task tools and reads the Markdown plans from planning mode. The spec is just a file. The tasks are just files. Everything stays editable and actionable.

Here's my new workflow: I write a short todo. Claude helps me turn it into a detailed spec, then into a task list, then works through the list. I check the kanban board, verify what's done, course-correct if needed.

I wanted to explore if we can build something like Claude Cowork in the browser that can work with the user's file system safely running programs against them. We can get a very long way: aifoc.us/the-browser-... Feedback and critique appreciated.

Looking forward to reading Maintenance: Of Everything by @stewartbrand.bsky.social How cool is this kintsugi cover

New flavor of imposter syndrome: I haven’t even used all my Claude Code quota today

5 years ago I switched to an ultrawide computer monitor and it’s been one of the best decisions I made. So much of my workflow these days starts as a single thread and then expands into 4 or 5 columns of side-by-side documents, AI agents and custom tools…

prototyping co-drawing with Gemini Flash 3 at Google in these demos "thinking" is disabled, which makes the model return tokens very quickly (all videos are realtime), and I find these rapid responses pretty good for the use-cases I'm experimenting with, like: executing simple diagrams ...

Wrote a blog post about the token encoding format in GPT tokenizers. Have you ever inspected a tokenizer and seen tokens like оже or ĠнÑĥжно? These are tokens in a custom encoding format used for serialization of byte-level tokenizers. This post shows how to recover the original token.