Dev tools with Jev

A Python command-line tool that asks Jev to classify, score or filter text and JSONL records. Local thresholds turn the answers into pipeline output and exit codes, making it useful for scripted checks and CI.

A local workspace for defining typed decisions, trying saved inputs and publishing fixed function versions. Backends access them over HTTP and agents over MCP, using scoped client tokens instead of sharing the vendor key.

Adds model-backed rules to Zod 4 schemas. A parse batches semantic questions into a Jev request, then translates threshold decisions into familiar validation issues; ordinary structural checks remain in Zod.

Places a relevance check in front of an existing test runner. Jev compares diffs with test source; local policy keeps uncertain tests and changed test files in the run. A model failure falls back to running the full suite.

A Flutter integration_test package lets developers register allowed actions and a step budget. Jev selects the next UI action from that set and evaluates goal completion, while Flutter performs the interactions.

A Rust CLI sends commit messages and diffs to Jev to label change types and potential security fixes. Results can be filtered by classification or CWE and exported for investigation; labels are review leads, not vulnerability confirmations.

A Spring WebFlux starter asks Jev whether a successful HTTP response actually contains an error or maintenance notice. Suspected failures become exceptions for existing error handling. If the model is unavailable, it defaults to allowing the response.

A Neovim plugin asks Jev a question about each Treesitter-extracted function in the current buffer and ranks hits in quickfix. It evaluates functions individually, so cross-function behavior needs separate inspection.

another demo for jev for ~instant variable generation

it's f*cking insane Jev sat between GPT and me, killing every draft that broke my rules before i…

love jev, but upset it - isn't open source? - it lack vision capability? We fixed all of…

e2e + jev from @typesafeai ⚡ I'm building an open-source framework for running e2e tests with agents. supports…

Yes, but can Jev do multi-agent swarms? Enterprise solutions that self-heal live productions before clients can even refresh…

Got early access to @typesafeai Jev and spent a few hours building on it. The problem I keep…

TYPESAFE RELEASED A MODEL THAT DOESN'T GENERATE TEXT, AND IT BREAKS THE COST FLOOR OF THE AGENT LOOP…

thanks to @typesafeai's Jev we now have massively parallel browser-based adversarial testing suite that tries to break each…

Jev unlocked a new way to stop AI agents from spinning. Agents do not always repeat the same…

Told my hermes agent to test the @typesafeai Jev model on an iOS app I’m currently developing for…

I build a SQL extension that turns plain english into a WHERE clause with jev @typesafeai WHERE jev(people,…

Inspired by @typesafeai , here is a Jev-compatible public API to play with It runs a comparable open…

视觉版 Jev 来了 仿照着社区的思路，把候选结果映射成固定标签，prefill 完直接采样，限制输出 token，采用多模态的 qwen-3.5-0.8B 4bit 量化部署在本地，16g m4 mbp 顺利运行 叠甲：这只是在 infer 层小小的复现一下 jev 的形式，肯定不如各种成熟推理框架效率高，也肯定不是 jev…

TypeGPU + ruNNtime + Jev @typesafeai is a very fun combo :D ruNNtime gives me efficient local inference,…

Built the ultimate syntax highlighter using @typesafeai Jev (i mean why not). It can highlight any language, even…

got Jev to review my PRs. ~200x cheaper than Claude and it answers in half a second 6…

I got access to Jev by @typesafeai today morning and I built a cool use case for it…

i have created moderation guard based on typesafe's jev its fast and accurate live demo you can:

Built an alternative version of @typesafeai but on @cerebras with Qwen 3.8 27b. Similar quality, similar performance, but…

built `jev-review` @typesafeai it's an experimental, local-first MCP plugin that gives coding agents a score quality feedback loop…

Testing apps with @typesafeai jev & @opencode will be soooo fast small demo:

On the Jev waitlist? You can use not Jev (but close enough) right here: browser demo

I increasingly think future software will combine code for control flow with small neural programs for "fuzzy" judgments.…