Case wall
Jev AI Agents cases, page 3
79 public posts filed under Agents. English summaries on this page, with a link to each case.
Last updated
These are the public posts filed under the Agents category.
No posts match these filters.
@Neriousy posts a short demo of testing apps with TypeSafe Jev and OpenCode, and says that combination will be fast. The post does not name the app under test.
Testing apps with @typesafeai jev & @opencode will be soooo fast small demo: https://t.co/WLSkpe9OZn
@moritzkremb says a quick test used Jev to improve a custom memory system, with 94% fewer tokens and 2–3x faster memory retrieval.
This is cool. I used Jev to improve my custom memory system. Results: → 94% fewer tokens used → 2–3× faster memory retrieval This was just a quick test, but I b…
@sawyerhood says Jev picks the agent, model, computer, and folder instead of filling those prompt-box fields. A major rewrite uses Fable plus Claude Code, and iOS app changes run on one of their Macs.
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 re…
@hqmank rebuilt a job crawler with Jev: from a company homepage it finds Careers and scores roles against their profile. They report about 5 minutes with an LLM versus just over 20 seconds with Jev in their test, packaged as the skill jev-job-hunter.
I rebuilt my job crawler with Jev. The task: start at a company's official homepage, find Careers, and identify jobs that match my profile. Before, with an LLM:…
@johnyeo_ says their Slack agent got 2x faster by using TypeSafe’s model to classify the best skill, tool, and params from the prompt before handing work to the agent, instead of the agent reading skills and choosing tools itself.
Jev made our Slack agent 2x faster ⚡️ Our agent can be quite slow because it needs to read skills and figure out which tools to call. We used @typesafeai's new…
@BhosalePratim swapped the LLM that decides tool calls for Jev. Their note is that an agent could act before the user finishes a sentence, and that voice agents might decide on partial utterances instead of waiting for the end of the turn.
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 wit…
@niazmorshed_ built jev-review, an experimental local-first MCP plugin. Coding agents call TypeSafe Jev while they work, get scored on quality metrics, improve, and repeat.
built `jev-review` @typesafeai it's an experimental, local-first MCP plugin that gives coding agents a score quality feedback loop across different metrics. age…
Quoted @CompleteSkeptic
@mdlahfir uses a deterministic hook so Claude Code can ask Jev where to delegate before it hands work off. They say mechanical tasks go to Haiku, intelligent ones to Opus sub-agents, and long-running implementation to an external harness (Claude Code, Codex, or Opencode).
Jev solved local harness/model routing I use a combination of Claude Code, Codex and Opencode as my local agentic stack and routing to other harnesses was alway…
@higgsfield_ai says their use case is auto-routing GenAI models: Jev picks the most cost-effective models for a task, balancing speed and quality. Generations in the demo are made with the Higgsfield API.
We found the perfect use case for Jev: auto-routing GenAI models. The idea: Jev picks the most cost-effective models for your task, balancing speed and quality.…
Quoted @higgsfield
@higgsfield_ai shows Jev plus Higgsfield routing generation models: Jev reads a prompt and picks models for video and image generation on the Higgsfield API.
Jev + Higgsfield = solved auto-routing for genAI models. In this demo, @typesafeai’s Jev evaluates prompt and picks the most fit models for video and image gene…
Quoted @higgsfield
@romdot2 demos pub-sub without a subscriber list: TypeSafe Jev does the routing almost instantly. A demo link is in the post.
pub-sub without sub @typesafeai jev does the routing almost instantly demo link below https://t.co/pTQIUtVBrI
@VladTerin built Jev Browser, an open-source adapter for existing Codex browser tools. The video is 1x speed: the user talks, Codex plans, Jev selects, and the browser moves.
I talk. Codex clicks. Watch the browser go⚡ This video is 1× !!! speed. Built Jev Browser: an open-source adapter for your existing Codex browser tools. Say the…
@mormonnegro wired TypeSafe and headless Chromium so agents browse faster. In the example the agent starts at the Spanish Wikipedia page for coffee and follows links to artificial intelligence in 20 seconds.
Acabo de terminar la implementación de @typesafeai + Chromium Headless para que mis agentes puedan navegar por internet a una buena velocidad! En este ejemplo l…
Quoted @mormonnegro
@stevekrouse posts a Jev plus Kernel browser-use demo and says it is fast. The try-it link is jev-browser-use.val.run; the post does not list the browser steps.
jev + kernel browser use demo pretty freaking cool! so fast! try it yourself: https://t.co/CoVFzscaa0 https://t.co/OyXdNA1H0F
@0x_kaize plugged Jev into Claude Code compaction as fast-jev-compaction. Instead of an LLM summary, Jev scores each tool call for importance, keeps what is above a threshold, and drops or stubs the rest, which they say avoids summary bloat and dropped context.
I got access to Jev and immediately plugged it into Claude Code to see if it actually FIXES the biggest pain of agent workflows: context compaction > result: pr…
@nidhisinghattri built a model-router CLI with Jev: you give a task and the AI subscriptions you have, and Jev says which agent or model should take it. They say pairing it with Herdr is the point.
i used jev from typesafeai to build a model router cli you give in a task and current ai subscriptions you have, the jev tells you which agent/model should pick…
Quoted @CompleteSkeptic
@princecaarlo runs an agent-browser and Jev loop for browser use. The stated goal is a Wikipedia race from Coffee to Artificial Intelligence.
okay so browser use is a legitimate use case agent-browser <> Jev loop goal: wiki race from 'Coffee' to 'Artificial Intelligence' https://t.co/565VqWYgoJ
Quoted @princecaarlo
@lomeshdutta open-sourced a Skill Router built with Claude and TypeSafe Jev. They have about 90 Claude Code skills and use maybe five, so the router is meant to surface a skill Claude would not call on its own.
Just hacked a fun little project - Skill Router (Open source) using Claude and Jev @typesafeai Why? I have ~ 90 Claude Code skills installed and use maybe five.…
Quoted @CompleteSkeptic
@iamzhe says minirouter/fusion is live: it classifies each request with TypeSafe Jev, then routes to a model matched to type and complexity. Codex lists their presets and routers in the model picker. They also fixed long prompts not caching automatically in Claude Code.
Small update for https://t.co/xXR4JLt377 today: 1. minirouter/fusion is live. It classifies each request with Jev by @typesafeai, then routes to a handpicked mo…
@kzkhykw used TypeSafe Jev to judge local Whisper transcripts and automate a Mac. Chrome pages use a browser harness; Slack and Notion are seen and operated through Mac accessibility. They say the implementation was done with Grok 4.6 only.
. @typesafeai のJevを使って、Local Whisperの文字起こしを都度判定して、Macbookのローカルコンピュータの操作を自動化して、あれこれするのやってみました！ Chrome のページは Browser Harness、Slack やNotionは Mac の Accessibility で認…
Quoted @kzkhykw
@0xkaushik_k built JevScope, which watches an agent and asks Jev whether each step is aligned, progress, repetition, or stuck, then plots that over time. They describe a coding agent stuck on a race condition where the curve shows it and no log line says “I’m stuck.” Code is said to be coming.
Got early access to @typesafeai Jev and spent a few hours building on it. The problem I keep hitting: agent traces tell you which tool ran, how long it took, wh…
Quoted @CompleteSkeptic
@Mnilax put Jev between GPT and themselves to kill drafts that broke their rules. Jev then stopped answering, and the agent treated silence as safer and sent nothing until a second agent unstuck it. They list branches for a quiet checker: pass it tagged unverified, hand the span to a big model, queue and retry, or wake a human if the action is irreversible.
it's f*cking insane Jev sat between GPT and me, killing every draft that broke my rules before i saw it. good setup. then Jev did not answer. the agent decided…
Quoted @Mnilax
@ashutoshftw built a semantic circuit breaker so agents that try different actions from the same wrong assumption still get stopped. In one stuck run they report 12 model calls down to 4 and 19.9K tokens down to 5.1K.
Jev unlocked a new way to stop AI agents from spinning. Agents do not always repeat the same action. Sometimes they try different things based on the same wrong…
@Mnilax wired Jev into the routing layer of a build that had been running on Astra and says the same task finished 37 minutes sooner with no malformed responses to retry. The post also cites TypeSafe’s published figures: $0.042 per million input tokens with free output, 67.8% matching Sonnet 5 on their eval at 294x less cost and 195x the speed, 0.114s versus 8.566s for GPT-5.6 Terra on the same demo, and about 10 decisions per second in the Doom demo.
man who co-invented ChatGPT just shipped a model that does in 0.114 sec what GPT Astra does in 8.5. i wired Jev into the routing layer of a build i was running…
Quoted @CompleteSkeptic