I did that once. Kept Opus going on a loop until Codex reviewer was happy.
It worked.
Probably.
I don't actually know.
The PR was over 3000 lines of code, fixing technical debt that had nothing to do with the PR itself.
Show more
My gut is that alot of AI output can be improved just by building a loop where a manager agent negs the other worker agent with prompts like:
"Are you sure this is the best you can do?"
"I think you can do better, try again"
"Take a closer look, give me 11/10 output"
Show more
Kind of a surreal experience:
- I open Linear
- Tell the Linear agent to do stuff
- I review the PR in Linear
- I open a preview build of Linear to test it
- I merge it and it gets deployed right away
- I open Linear
Show more
Linear's coding sessions now supports custom environments and browser use for testing and screenshots.
Configure runtimes automatically with agent, and setup necessary scripts and env vars.
Show more
We updated diffs in @linear. You can now preview images, SVGs, and GIFs, compare changes side by side, adjust opacity, and view sources.
We’re still pushing to make diff reviews faster and nicer to use. Give it a try we’ve shipped a lot of improvements lately.
Show more
I added this GitHub incident badge during my first weeks @linear as a quick-win, thinking no one will probably see it.
Boy, was I wrong...
We added zig syntax highlight to @linear.
Idk why we didn't have it yet.
People bring up non-determinism of LLMs forget that the modern computing have to deal with quantum effects in MOSFET transistors, thermal noise, cosmic rays, fluctuating power supply, silicon defects.
We just gradually figured out how to make them super reliable.
Show more
100%. Higher level code, frameworks and libraries are for humans and they all are on their way out.
The reason I love working @linear is that this feature took 0 meetings, 0 RFCs, no A/B testing, no alignment with PM, no cross-team alignment, no H2 planning, no committees, it moved no KPI (as far as I am concerned). It felt good so we built it.
Show more
Look, sometimes you write the code first and _then_ backfill an issue.
There's now a handy button for that!
You can now hold Cmd and click on code references in @linear to jump to their usages or definitions.
Crazy how Maestro itself went from an experimental feature last year to automatically ordering latest iPhone on Chrome
And just like that...
Maestro is now working on Web!
Check out the experimental web driver in action below:
Today we are excited to announce the next evolution of Maestro Studio which brings maintainable no-code UI automation to the Mobile ecosystem! 🎉 🎉 🎉
🧵
Show more
This has been a long-requested improvement and I am so proud of the team for finding such an inventive way of addressing it
If you've used Maestro for iOS in the past, you may have encountered issues matching certain UI elements.
We've spent the last month re-architecting the driver to solve these issues for good 🙌
Today, we are ready to announce Maestro's new iOS Driver!
Show more
If you've used Maestro for iOS in the past, you may have encountered issues matching certain UI elements.
We've spent the last month re-architecting the driver to solve these issues for good 🙌
Today, we are ready to announce Maestro's new iOS Driver!
Show more
What amazes me about this feature is that it was built in under a week and yet it is clean and very much maintainable. Kudos @LelandTakamine !
Introducing: Maestro Studio 🚀
Instantly discover the exact commands needed to interact with your app!
Meet your new favorite Maestro tool ↓
🔥 Introducing: Maestro Cloud 🔥
The easiest way to run your Maestro Flows on CI
Read the full announcement here:
🧵
📢 BIG Maestro update today (v1.10.1) 📢
Let's dive right in! Starting with.... 🥁
🎉 Conditional Flow Execution! 🎉
In the video:
1st run - ✅ Condition succeeds + nested flow is executed
2nd run - ⚪ Condition fails + nested flow is skipped!
1/7
Show more
Exciting announcement for Android devs:
Maestro 1.9.0 now supports USB-connected Android devices! 📲
🙌 🙌 🙌
Upgrade to the latest maestro CLI to try it out!
And please let us know you run into any issues or have any questions. Happy testing!
Show more
