What people are shipping with Jev.

JevNoiseGate filters unwanted notifications and SMS on Android.

@fernandoeeu_dev @typesafeai A gente tinha isso determinístico antes. Mas errava muito. Aí a gente aumentou o limite do determinístico pra trazer uns 20 itens e o jev tá decidindo

shipped: jev by @typesafeai now gates tryeve's builds.

got early access to @TypeSafeAI (Jev). in my last post on autoresearch vs dream-rsi, the quiet bottleneck was: how do you score 28+ dreamt tree branches offline without waiting mi

Found the perfect use case for @typesafeai Jev: instant compaction in 2026. Why is compaction still a summarization prompt? Jev scores every tool call and drops what’s irrelevant—m

@ThisisMarkT @typesafeai Yes, as a filter or reranker rather than the generator: retrieve wide, ask Jev per chunk "does this answer the question", keep only the high-probability on

I think @typesafeai Jev has solved a much bigger AI problem than it first appears: LLMs shouldn’t have to think about everything. A huge amount of “reasoning” is actually classif

@alexatallah @typesafeai low-entropy routing only pays if the classifier is cheaper than sending the PR review to the frontier model. are you actually splitting those three buckets

Got access to jev by @typesafeai so wanted to test it out asap. At @parsewise we run topic relevance checks as a filtering step on our data processing pipeline. A task that is pr

I am going to build for companies who have lot of candidates to evaluate based on resume I am using the coolest model JEV by @typesafeai to analyse all of them and find the rig

@chaturvedikun @typesafeai bfs for shortest path then jev on spatial metrics is a clean handoff. decision models belong on the branch not the path search same split i used for offl

La rapidité pour les recherches web en intégrant jev de @typesafeai est incroyable 🚀 Quelques secondes pour des dizaines de recherches sur @okazfr. Et ce n'est que le début... ht

@ethereaglehq @o_kwasniewski @typesafeai both are valid, different layers. Jev is System One: Choice/Score/Noul over state, not a browser driver. so "slot in as pass/fail (or sever

instead of generating text, jev from @typesafeai generates structured output this makes it great for classification tasks like model routing, tool selection/search, and guardrails

Built a support-ticket triage simulation with @typesafeai #Jev ⚡ 5 typed judgments in parallel: • Department • Frustration • Urgency • Human request • Refund intent ~126ms avg |

Jev from @typesafeai cannot write sentences but is great at deciding and classifying This makes it a direct alternative to LLM-as-a-judge setups. Jev returns a choice between pr

@langfuse @typesafeai @langfuse The certainty signal is what I'd push on. How does Jev behave when confidence is low but it still has to return a category? That boundary between "u

Switched a big chunk of matching, routing tasks to Jev by @typesafeai Crashed my bill by 87% I was already using DSV4F btw Crazy times we're living in

Jev for first pass, LLM for the gray zone @typesafeai ftw https://t.co/tkVoCWYQ2U

@barckcode Hay ya decenas de enrutadores de modelos con Jev. Es un caso de uso bastante claro. https://t.co/1GcFAm02v6