What people are shipping with Jev.

been messing with using jev to help an ai find the right stuff in its old history instead of cramming everything into one prompt got 12/12 in a small experiment, even knew when t

thanks to @typesafeai jev I no longer have fill out all of those fields on prompt boxes. It picks the agent / model / computer / folder for me. - For a major rewrite it uses Fab

@MicahDail @netestapper @typesafeai A bunch of agents are also slapping together an experiment of a kind of Jev loop where, during a /goal, the agents consult to be able to evaluat

My 11 year old earned a Bambu Lab X2D. It landed a week before GPT-6 Astra. Astra gave us CAD for whatever he wants to build. Then @typesafeai shipped Jev, so the robot can decide

Played around with @typesafeai Jev today, mostly to understand what it does for tool calling. Instead of an LLM deciding what to do, I substituted that part with Jev. My learni

Same week TypeSafe launched Jev (@typesafeai / @CompleteSkeptic). Not a chatbot. Send state + typed questions. Get decisions with probabilities. The pitch is that it's very fast a

5 ways I’m going to bolt JEV onto my DGX Sparks and see how far I can push local agents fleet. 👇 JEV might be one of the most interesting things Ive found for my DGX Spark fleet

Muse can go through all of Facebook and Instagram to find potential customers for your products. That’s a ton of data so I’m experimenting using Jev to see how good it is at quickl

we're using Jev in our robotics system AMA

@typesafeai Post 4/7 The biggest gain came from a traffic-policy incident: 0/5 → 3/5. Baseline agents chased plausible telemetry noise. Jev rejected a weak causal explanation and h

@K_aditya25 @typesafeai the latency win is less "Jev does browser" and more "stop paying an LLM to plan in prose". kyle's loop is the pattern: observe a11y as state, Choice over ca

@typesafeai @CompleteSkeptic System One models are emerging: no chat, just decisions, returning structured values with probabilities. If these mature, agent workflows will re-layer

@o_kwasniewski @typesafeai another builder picking jev as the agent layer for e2e. the dom snapshot approach keeps beating screenshot loops on reliability. what handles the mobile

Sure glad @typesafeai made Jev, so I can get back to work while my AI plays SSBM with its buddies. https://t.co/Xhom1l59Yb

@CryptoPraetoria @ownerofmarket @AsideAI @typesafeai Jev is better for browser navigation and execution. It chooses only from real page elements with typed decisions in ~70-500ms,

@rauchg @typesafeai @vercel Default-auto with a safety reviewer on every command is the agent posture I want more CLIs to copy. @rauchg on @typesafeai fx (auto mode, reviewer on G

@NavDoesTech @AsideAI @typesafeai not quite. nothing about the site needs to be known beforehand. the browser hands you the real candidate set at runtime. that's what @AsideAI is b

@frmdomi @AsideAI @typesafeai if u mean agentic = generative, yes, jev can't write prose. but agency isn't generation, it's the observe decide act verify loop, and that's literally

Jev by @typesafeai turns your  Mac into an AI agent... what this means for You ? - Jev, a new model from a ChatGPT co-inventor, can operate computer tools in real time. - It is

@pdp @typesafeai Decision API as a first-class surface is the right packaging. Agents keep the generator; /v1/decision/create owns the gate. Curious — are you routing tool choice