Catherine Kim

16 posts

Catherine Kim

@inkycat

intern @imbue_ai | fan of puzzles, musicals, and social deception games/shows

Beigetreten Şubat 2026

13 Folgt19 Follower

Catherine Kim@inkycat·1d

These games are amazingly fun to play and very creative! props to the team and super excited to get working on arc-agi-3

ARC Prize@arcprize

Announcing ARC-AGI-3 The only unsaturated agentic intelligence benchmark in the world Humans score 100%, AI <1% This human-AI gap demonstrates we do not yet have AGI Most benchmarks test what models already know, ARC-AGI-3 tests how they learn

English

10K

Catherine Kim@inkycat·1d

@imbue_ai @modal never give up on a repo because you can't get it running again 💪

English

Imbue@imbue_ai·1d

Teach your repo how to run itself 🦾💨 Introducing Keystone: a self-configuring agent inside a sandboxed @Modal container that generates a working dev container for any repo → pip install imbue-keystone

English

13.5K

Catherine Kim@inkycat·1d

--dangerously-skip-permissions with ease and grace

Imbue@imbue_ai

English

220

Catherine Kim retweetet

Imbue@imbue_ai·3d

Offload just demolished xdist on our test suite

English

1.1K

Catherine Kim@inkycat·3d

@nikhilville ur welcome king 👑

English

Nikhil Kulkarni@nikhilville·5d

lunch today hit

English

Catherine Kim@inkycat·19 Mar

@imbue_ai @modal LOVE how fast it makes tests run

English

163

Catherine Kim retweetet

Imbue@imbue_ai·19 Mar

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

194

99.5K

Catherine Kim retweetet

Imbue@imbue_ai·11 Mar

…the average review takes 20mins and $25 💀 We made a better alternative. It’s called Vet and it’s basically free. It runs in less than 30 secs (up to 5 mins for full PR reviews), only costs inference, and it can run in the loops agents work. We open sourced it: github.com/imbue-ai/vet

Claude@claudeai

Introducing Code Review, a new feature for Claude Code. When a PR opens, Claude dispatches a team of agents to hunt for bugs.

English

1.6K

Imbue@imbue_ai·5 Mar

Your coding agent may be lying to you. You ask it to write tests. It says they pass. It never ran them. You ask for a feature. It hits a wall and silently swaps in fake data. We built Vet to fix this. It’s open source. Get the code below.

English

378

142.5K

Catherine Kim@inkycat·5 Mar

@imbue_ai finally don't have to parse through all the reasoning to figure out if it lied 🙏🙏

English

110

Catherine Kim@inkycat·28 Şub

@thisisehsan @imbue_ai WORKING ON IT!! STAY TUNED FOR MORE IMBUE LAUNCHES 🎉

English

Ehsan Nour@thisisehsan·28 Şub

@inkycat @imbue_ai LEGEND INTERN STATUS yet? or?

English

Catherine Kim retweetet

Imbue@imbue_ai·27 Şub

Today we’re open sourcing Evolver, a near-universal optimizer for code and text. While benchmarking we achieved SOTA (95%) on ARC-AGI-2 (last week that is 😆) and 3x’d performance of the best open model, reaching GPT-5.2-level performance.

English

104

937

127.2K

Catherine Kim@inkycat·27 Şub

@nikhilville lowk i hate bananas but the banana milk was too good

English

Nikhil Kulkarni@nikhilville·25 Şub

I take one sip of the banana milk and immediately I'm locked in

GIF

English

Catherine Kim retweetet

Imbue@imbue_ai·27 Şub

We built Evolver to improve our open source agentic code verifier, Vet. In other systems, we use it as an automatic prompt optimizer. How Evolver works: imbue.com/research/2026-… How Evolver set a record on ARC-AGI: imbue.com/research/2026-…

English

108

11.3K

Catherine Kim@inkycat·27 Şub

@imbue_ai research is goated here 💯

English

483

Entdecken

@imbue_ai @modal @nikhilville @thisisehsan @elonmusk @BarackObama @taylorswift13 @cristiano