I’ve been trying vanilla Codex with remote control running on a MacBook 24/7. Computer use enabled, browser with logged in sites. Giving it tasks from my phone.

It’s been nice overall, but Astra (or its harness?) isn’t particularly proactive and autonomous. You have to kick it in the butt for every step. Written it up here: https://ivan.berlin/notes/chatgpt-personal-assistant/

Looking forward to this new Duet thing!

Any reason you're not using ChatGPT Dots which is also using Astra (and also unlimited free usage this month) and precisely designed to solve the proactive/autonomous issue? https://openai.com/index/introducing-dots/

I would love to, when they are available in EU.

What a beautiful font! Thanks so much for open-sourcing it!

Slowly reading “Seeing that Frees” by Rob Burbea.

Highly recommended if you want to challenge your view of the world.

Meetup.com is terrible and we definitely need a replacement. But this is not it.

https://www.LokiList.com is more in the spirit of Craigslist casual encounters but it might also be used to plan strictly platonic group activities.

Fair enough - appreciate the feedback either way :)

What would make it “it” for you? Is there a specific feature(s) you think is missing?

Nice project!

Here’s my setup: https://mysetup.ai/u/katspaugh

Mostly vanilla Claude but inside a VM.

What’s great about it? I shudder every time some product manager sends out a miro link.

People should try Kinopio.

It's a best-in-class real time collaboration board. It's also a decent general purpose diagramming tool, but if you're using it for creating software architecture diagrams or product flows, there are better tools for that.

Whimsical is my favourite tool for that.

Miro was always fine to use at work. It’s no Visio desktop or OmniGraffle but it wasn’t offensive and I only somewhat preferred LucidChart over it.

it was excellent at whiteboarding remotely.

The anthropomorphic delusion about LLMs being a coherent entity, with memory and continuity (which is also a delusion even when it comes to humans), quickly evaporates when you realize that each message in a chat is a completely separate stateless API request, and the only trick fueling the illusion is that the previous messages are sent to the server along with your new message. So AGI or not, it’s just a process that exists literally for the duration of one API request.

I've written two QA skills for our app: 1) flow discovery, and 2) e2e implementation.

The first one scans the app and detects critical flows, happy paths, unhappy paths, invariants. Writes an invariant matrix across app domains and surfaces, and maps acceptance criteria to each.

The second actually implements the flows as e2e tests (Playwright for web, Maestro for mobile). Tests are run locally before a PR is submitted, with a video recording. The video + screenshots are posted in the PR body. A rule in CLAUDE.md says "No PR w/o visual evidence!".

CI runs the suite depending on touched code.

It's nice to lay back and watch the runner click through your app.

Cute font!

Round, like we’re trending towards Comic Sans but still professional.

Looks cool!

I took a similar approach with https://runmachine.dev/ but later switched to OrbStack for iOS development.