Krishnan

agent babysitter

[Krishnan @krishships]

Ok, i'm a few months late to the background agents thing.

I didn't want to build another one you use solo. Building software is a team activity, so working with an agent should be too.

The whole idea behind Codevil is keeping your team in the room with it.

It's built around shared sessions. Plan something out with your designer, brainstorm with the eng team, kick off a session, pull in a coworker to help debug, hand it off when you're done. Everyone's in there with the agent.

It all runs in the cloud. Every room spins up its own sandbox, clones your repo, and gives you a live preview the whole team can watch. Built on Pi + Cloudflare (Workers, Durable Objects, Sandboxes).

[andre --dangerously-skip-permissions @andrezfu]

Agents write application code, but you still get paged at 2am when it breaks.

We wanted to know whether AI could handle that part of the job too.

Introducing Incident Arena: a benchmark that puts coding agents on call!

Check out our paper & full dataset release below!

[Krishnan @krishships]

others: using Instinct to be more productive, offloading chores and ofc do restaurant reservations

me: using Instinct to convert my meme collection into WhatsApp sticker

[Lance Martin @RLanceMartin]

i tested sonnet 5.5 on "code-to-painting". every pixel is generated by python the model wrote. it's shown a photo, writes a brush engine, renders, looks, and revises. you can see the large step-up from sonnet 5 in terms of coding + visual reasoning.

[Krishnan @krishships]

Gave AIyappa access to Langfuse so he can debug and improve on his own.

tbh Slack cooked with blockit, makes rendering a variety of ui blocks so easy.

[Krishnan @krishships]

Created my own version of Claude Tag and named him AI-yappa

[ben hylak @benhylak]

who did this

[Krishnan @krishships]

Created my own version of Claude Tag and named him AI-yappa

[Krishnan @krishships]

neighbours got no chill

[J @writer4lyf]

Feel like BKU should be running in theaters year long so whenever you feel low you can just pop in and watch it.

[Krishnan @krishships]

@ankitkr0 Morph?

[Krishnan @krishships]

@pHequals7 @adeatii https://t.co/hJiusxazr6

[Gergely Orosz @GergelyOrosz]

Went to NVIDIA HQ today.

Two interesting observations:

1. Snacks and coffee are not free: you have to pay for them. This would be unusual at Big Tech, but no big deal for devs here. "We use this thing called salary to buy stuff we actually need." Food for thought (literally!)

[Aporia @aporia9n]

I think ambitious people should treat life in “seasons” rather than trying to balance everything every day.

A few years massively overallocated to work. A period getting seriously fit. A year or two nomading. Then family takes priority for a while. Then maybe at 35 you go completely insane again and spend four years building a new company, and so on.

The mistake is thinking every metric needs to stay green all the time. Most great outcomes (building a company, writing sth serious, training for an event) need uninterrupted runs, not daily moderation. Sometimes work should suffer cause you’re travelling, sometimes your social life should suffer cause you’re building, sometimes career progression should slow cause family matters more. That’s fine.

And tbh it probably makes life more fun too. You get actual chapters and completely different versions of yourself instead of spending 40 years maintaining the same perfectly optimised routine. A few intense years in one city, a weird nomad phase, an obsession that takes over your life, then something completely different. Probably much more memorable than one very long well-optimised Tuesday.

It's why I encourage everyone to think of balance in years, not days. A good life can look horribly unbalanced on some random Tuesday and still be very balanced over 10 years.

[7/11 Truther @DaveMcNamee3000]

Me a 32 year old unemployed man walking into the 11 am screening of minions

[Krishnan @krishships]

Nice try Comet, but Theepori Thirumugam did it first

[Krishnan @krishships]

@CloudflareDev Don't have a publicly accessible deployment yet but building a real time collaborative coding workspace where teammates and AI agents work side by side to brainstorm, implement, review, and ship changes.

Built completely on top of Cloudflare primitives
https://t.co/F4bomtzDmY

[Krishnan @krishships]

Ok, i'm a few months late to the background agents thing.

I didn't want to build another one you use solo. Building software is a team activity, so working with an agent should be too.

The whole idea behind Codevil is keeping your team in the room with it.

It's built around shared sessions. Plan something out with your designer, brainstorm with the eng team, kick off a session, pull in a coworker to help debug, hand it off when you're done. Everyone's in there with the agent.

It all runs in the cloud. Every room spins up its own sandbox, clones your repo, and gives you a live preview the whole team can watch. Built on Pi + Cloudflare (Workers, Durable Objects, Sandboxes).

[Krishnan @krishships]

honestly it’s nice having limit resets I can use whenever it suits me. codex ftw.

[Krishnan @krishships]

@thdxr Gangprompt with Codevil soon

[Krishnan @krishships]

Ok, i'm a few months late to the background agents thing.

I didn't want to build another one you use solo. Building software is a team activity, so working with an agent should be too.

The whole idea behind Codevil is keeping your team in the room with it.

It's built around shared sessions. Plan something out with your designer, brainstorm with the eng team, kick off a session, pull in a coworker to help debug, hand it off when you're done. Everyone's in there with the agent.

It all runs in the cloud. Every room spins up its own sandbox, clones your repo, and gives you a live preview the whole team can watch. Built on Pi + Cloudflare (Workers, Durable Objects, Sandboxes).

[Krishnan @krishships]

@_mctrinh A cloud coding agent with collaboration at its core

[Krishnan @krishships]

Ok, i'm a few months late to the background agents thing.

I didn't want to build another one you use solo. Building software is a team activity, so working with an agent should be too.

The whole idea behind Codevil is keeping your team in the room with it.

It's built around shared sessions. Plan something out with your designer, brainstorm with the eng team, kick off a session, pull in a coworker to help debug, hand it off when you're done. Everyone's in there with the agent.

It all runs in the cloud. Every room spins up its own sandbox, clones your repo, and gives you a live preview the whole team can watch. Built on Pi + Cloudflare (Workers, Durable Objects, Sandboxes).

[Krishnan @krishships]

@phalgooon Vibe coders test in prod so...

[Krishnan @krishships]

Ok, i'm a few months late to the background agents thing.

I didn't want to build another one you use solo. Building software is a team activity, so working with an agent should be too.

The whole idea behind Codevil is keeping your team in the room with it.

It's built around shared sessions. Plan something out with your designer, brainstorm with the eng team, kick off a session, pull in a coworker to help debug, hand it off when you're done. Everyone's in there with the agent.

It all runs in the cloud. Every room spins up its own sandbox, clones your repo, and gives you a live preview the whole team can watch. Built on Pi + Cloudflare (Workers, Durable Objects, Sandboxes).