Ainesh Chatterjee

MTS @FactoryAI. I code often and sleep occasionally. I like systems. Agents are graphs. Harness is everything. Views are my own. #2 fan of Corvus Corax

[vx-underground @vxunderground]

Microsoft introduces Microsoft Scout, also known as Autopilot.

Scout is always on and has file system and application access "based on your corporate policy".

Best news for Threat Actors in a long time

https://t.co/M3pyfcbTBm

[Ainesh Chatterjee @ain3sh]

@badlogicgames Heh pipi 🙂‍↕️

[🚨 AI News | TestingCatalog @testingcatalog]

@OpenAI How I feel when AI News arrives 👀

[0xSero @0xSero]

8-16gb vram bros rejoice! You have a new best in class model ❤️

Thank you Google

[Ainesh Chatterjee @ain3sh]

@delba_oliveira Ye, model tasks like what it’s seen in RLVR

[Ainesh Chatterjee @ain3sh]

@lumendriada Same hook should work in droid too

https://t.co/dMuzqLCJO1

[herdr @herdrdev]

new herdr update!

agent session restore is now enabled by default for supported integrations.

after a herdr restart or update, supported agents can automatically resume their native sessions right where they left off.

also introducing release channels. a small thread:

[Emil Kowalski @emilkowalski]

To get good animations from an AI you need to get good at telling it what you want:

- "stagger this list of items"
- "make this animation direction-aware"
- "spacial consistency", "crossfade", "layout animation",

I made a motion vocabulary for this:
https://t.co/ExAxpr31no

[Brotzky @brotzky]

Conductor made their app 2x faster.

I sat down with @jacksondecampos to learn exactly how they did it.

Here's the story of the rewrite:

https://t.co/7zYxHZJqWR

[Jerry Liu @jerryjliu0]

This is actually one of the main advantages startups have over frontier labs, as long as there's a healthy spectrum of open-weight to closed-weight models on the cost-performance curve.

Building a model harness is fundamentally a bit different than building the underlying model. The last year has shown that an increase in model capabilities has led to an increase in sophistication around agent harnesses as well (e.g. just look at Claude Code ultracode)

Every startup has a meaningful advantage here - from Factory to Cognition to honestly us - if they're able to optimize and finetune the model set for the precise amount of intelligence needed to do a task. Same or higher accuracy at 2-10x lower cost and latency is a real advantage.

[Factory @FactoryAI]

Introducing model routing to Factory.

Factory Router picks the right model for every task, automatically.

Maintain frontier performance while cutting costs by 25%.

[dax @thdxr]

someone made a fork of opencode that routes through the unsecured ai endpoints from chipotle

[Francesco @francedot]

1/ This week our team at @trycua worked hard to ship cua-driver v0.5. The headline is a new primitive, Agent Cursors: many background agents can now each own a cursor, scoped to a non-overlapping action space.

[τop τick crypτo 📁 🤖🧠 @toptickcrypto]

“The memory trade can’t keep going because the hyperscalers can’t spend more than 100% of cash flow”

[Ainesh Chatterjee @ain3sh]

@dexhorthy Default behaviour in droid 😼

[Florian Brand @xeophon]

For safety reasons, we made the model worse at <thing>.

You can, however, contact sales and get exclusive access to the model that’s even better at <thing> for 10x the price.

[Browser Use @browser_use]

Introducing: New Browser Infrastructure. ⚡️

Chromium fork. Firecracker fork. Custom Linux kernel.
> 3x cheaper. $0.02/h.
> Unlimited scaling
> Subsecond cold starts
> Most stealth browser infra

Feel the speed. Now live.
Try it now ↓

[Ainesh Chatterjee @ain3sh]

@TelepathicPug @50onice 🥹

[Ainesh Chatterjee @ain3sh]

@kmcnam1 Wait wut