What people are shipping with Jev.

Il @typesafeai Jev works properly, I think it could really change/improve multiagent implementation. Let me explain:

Built a small AI customer-triage simulation system with TypeSafe’s Jev - @typesafeai Instead of asking an LLM to generate a long response and parsing it, instructed Jev answers 5

I've had access to Jev by @typesafeai for about 30 hours now and I can't stop thinking of the million places it can be used 🤯 Meanwhile, checkout this fun snake game I built wher

@LegendrSaiyajin @joshbuildings @theo @kwuchu @typesafeai The LLM wrote the harness talking to jev, instead of talking to jev it could’ve just written the code to control the playe

@mvanhorn @typesafeai I made GLiNER do the same agentic decision making tasks like Jev while being local, free and open weights. it's so lightweight it'll run on a laptop. https://

@typesafeai need jev access :)

@typesafeai Wanna save your old boss @OpenAI millions of $$$? Tell them to implement ‘JEV’ for age verification (from user data) and use Persona as a last resort when ‘JEV’ cannot

Your repo. Your models. Your tools. QQ is an open-source terminal agent with Codex sign-in and an optional JEV companion for evidence reviews. We're live on Product Hunt—would lo

jev is insane 🤯 Same Combos FPS. Astra left. @typesafeai Jev right. Reasoning model vs decision model — not even close. Jev does not replace astra, it opens a different game. http

I’ve put a ton of tokens through Jev now, and...I have thoughts. 1. The big labs lost the script TypeSafe’s intro video says, “We’re building prod, not God” — and can I just say?

Show HN is up if you want the boring production layer around Jev: policy + confidence gate + shadow + recipes + action-level evals Measured on our machine: Claude CLI 48.9s → Jev

@MiaAI_lab Would love to see any results from your findings! As an experiment, could Jev help quickly compact small context windows like: https://t.co/L3O4WPEJMJ

@drewpb123 @chesscom Feel free to play with Jev here https://t.co/e36Tx1HOYK Source: https://t.co/jhhVZ89cd0 https://t.co/l85m3q8N4l

判定特化AI「Jev」に将棋を指させてみた。 Jevは文章を生成せず、選択肢ごとの確率だけを返すモデル。毎手、合法手ぜんぶを渡して1手を選ばせた。 結果は将棋エンジンの最弱・弱に勝ち、中以上には負け。5局で約$0.035。 ロリポップ！AIゲートウェイ経由で今週無料です！ https://t.co/2oni7X6gOD https://t.co/i7u

Jev Agentめっちゃループしちゃう... browser useとかはどう実装してるんやろ 今日はもう寝る https://t.co/WZi0Vm5YCc

Two commands inside Claude Code: /plugin marketplace add EliaAlberti/jev-rules /plugin install jev-rules@jev-rules Repo, docs and the full-quality demo: https://t.co/kFefHnJJSa

@studio_yebisu I can’t see mine! https://t.co/kFefHnJJSa Jev picks which of your rules apply to each prompt, so Claude only sees the ones that matter.

Put @typesafeai's Jev in front of your LLM. RAG gate → keep only high-confidence passages. Tool gate → allow / escalate / block before refunds, email, deletes. Typed decisions fir

Introducing Jev-is-odd You can ask Jev directly if a number is odd or not AGI is here https://t.co/3vj9EqU7D9 https://t.co/1rPq5eFM3N

Kody v2026.09.18 is out. Jev Score rerank for ranked search. When the jev-search-rerank flag is on, ranked list-mode search widens its hybrid pool (~50 candidates), then a second s