Roy

Building…

[Roy @Roy__Gross]

https://t.co/dYqoflq8gC

[Roy @Roy__Gross]

Started playing with Claude Code routines and found the "auto fix PRs" setting.

I define the loop → Claude opens the PR → agentic reviewers leave comments → Claude fixes or ignores them → it waits for checks to pass → then I get a Slack notification that it’s ready for me.

Pretty close to fully agentic development, as @bcherny intended

[Roy @Roy__Gross]

@mktpavlenko @linear You can fully customize what gets included, so you can remove comments or attachments if they feel like noise - I personally found the extra context useful for the agent

[Roy @Roy__Gross]

Linear’s "Copy as prompt" is such a quiet power feature.

One click and I get the parent issue, comments, attachments, and a clean customizable prompt. I just paste it into a new workspace and the agent already has full context.

Really well done @linear.

[Roy @Roy__Gross]

Tried Claude Code's routines (loop engineering) and it's actually impressive.

I set up a daily loop that finds feature flags marked "ready for cleanup", removes their usages, and opens a PR.

In one week it cleaned up 25 of them. Feels great 😄

[Roy @Roy__Gross]

@benln Add me

[Roy @Roy__Gross]

@peer_rich @DBredvick A worktree can be dirty if the infra is shared (db, files, queues, etc)

[Roy @Roy__Gross]

@jonathanfishner Sounds cool, do you support bring your own subscriptions?

[Roy @Roy__Gross]

@FlyaKiet Maybe a new agent escaped the sandbox? 🙃

[Roy @Roy__Gross]

בתזמון מושלם

[Cursor @cursor_ai]

Origin, our code hosting platform, is now live.

It's fast, easy to use, and deeply integrated with Cursor.

Get started by syncing your repos from GitHub.

[Roy @Roy__Gross]

@eliron_giny לא הייתי מגדיר נכון לכרגע את מהפכת הai כ״התקדמות בחיים״ אבל בעתיד כנראה שכן

[Roy @Roy__Gross]

And now we wait

[Roy @Roy__Gross]

@willcb Linter >>> skills

[Roy @Roy__Gross]

@yonyoniz בלי סוכר 🫠

[Roy @Roy__Gross]

@Assaf_Nathan אני משתמש בשנים האחרונות ב Saily שהוא מעולה , התקנה חד פעמית ככה שכל מה שצריך זה רק לרכוש את החבילה למדינה כשרוצים.. אולי יעלה כמה שקלים יותר אבל שווה את זה

[Roy @Roy__Gross]

@FlyaKiet @superset_sh Non-solo devs don’t care about token spend.. they all use the company’s subscriptions/ tokens

[Roy @Roy__Gross]

My whole feed is @bot 🚀

[Roy @Roy__Gross]

@jjacky Local first approach with a container for each workspace/worktree

[Roy @Roy__Gross]

רדיפה אחרי כל שורת קוד שנכתבת כדי לצעוק ai slop זה כבר לא באופנה

[Santiago @svpino]

I'm officially done reading AI-generated code.

It's been two weeks since I looked at any of it.

I think the IDE is officially on its way to the graveyard. The job is no longer about "writing code," so we need new tools that better reflect this new reality.

While reviewing the code, I realized my only complaints were stylistic, and I wasn't finding any obvious bugs anymore.

The more code I generated, the harder it became to keep track of every line. I found that my time is better spent designing ways to verify that the overall system works than looking at the code.

State-of-the-art coding agents are better at writing code than I'd ever be, and I'm going to stop pretending otherwise.

I still think these coding agents can't go too far without an experienced human guiding them, but we're past the point where we need to check every line of code.

[Roy @Roy__Gross]

@ronbmp שיחקתי קצת עם muse 1.1 כשהם חילקו 20 דולר קרדיטים כשהוא יצא והוא היה אחלה - אני מנסה לאפטם דרך subscriptions כל עוד הם קיימים ולא דרך api usage שיכול לשרוף בשעה מאה דולר