Jev's Picks

Popular posts about Jev on X, best ones first

13.7KPosts
402.6MViews
157.8KReplies
209.5KReposts
2.3MLikes
何笠同学
@heli8866

😱 Holy moly, the value of this trick is absolutely off the charts!! Anthropic recently released a really handy method in Claude Code. ⬇️ The idea is to run three models at the same time: Sonnet, Haiku 5.5, and Opus 5.5, each handling only the work it does best. The results are better, and tokens are cut down a lot too, because every task goes to the most suitable model. When launching Claude Code, just write it like this: claude --model sonnet --advisor opus The division of labor works like this: → Sonnet 5.5 runs the main session, writing code and fixing bugs → Haiku 5.5 works in parallel inside the repo: searching, grep, reviewing diffs → Opus 5.5 stands by as an advisor, handling only key decisions and signing off on the diff at the end → Tasks that don't need Opus shouldn't waste Opus's context window Opus only steps in at three points: → Before the plan is settled → When the same error shows up a second time → Before declaring the work finished Sonnet writes, Haiku checks, Opus reviews. JEV engineering takes the same idea one layer further: the forks that don't require thinking (which file to pick, which tool to use, retry or stop) are handed to JEV, which handles them in a few milliseconds, and only the decisions that truly require judgment go to the large model. Full structure: > Sonnet 5.5 runs the main session and coordinates the overall work > search uses Haiku 5.5 to map out the repo, read-only > grep uses Haiku 5.5 to locate call sites, with an independent context > audit uses Haiku 5.5 to check diffs, running in a temporary worktree > Opus 5.5 stands by as an advisor and reviews the consolidated diff before sign-off > JEV routes all the small forks in milliseconds Paste this structure along with the following prompt into Claude Code ↓ "Reconfigure my Claude Code according to this structure: 1. Use Sonnet 5.5 for the main session and Opus 5.5 as the advisor. 2. In .claude/agents, create three Haiku 5.5 subagents: search (map out the repo), grep (locate call sites), and audit (check diffs). 3. Set isolation: worktree for audit so it doesn't touch my working branch during checks. 4. Add a rule to CLAUDE.md: all repo searches and tests are delegated to the Haiku 5.5 subagents to keep the main context lean. 5. Hand fast, deterministic forks (tool selection, path selection, retries) to JEV. For every change, show me the diff first, and don't modify any files until I say to start." 🔗: Recommend bookmarking this, you'll definitely need it later.

Quoted on Jev-X by @yakumo
Brennan Dunn
@brennandunn

It's cheaper (like, $1.50/year) to have Jev just regularly analyze whether our garage door is open or not vs. actually ordering and installing a reed switch. Home Assistant orchestrates, and if it's left open unintentionally I'm pinged. Real world "hot dog, or not hot dog?"

Sahil Panhotra | Indie Builder
@sahilpanhotra

Man, reading posts like this is depressing. Finding a new job shouldn't become another full time job. I'd love to just give an AI agent my skills/resume and let it find relevant openings, compare salaries and find people worth reaching out to. You don't even need a subscription for this. $0.03 for a full job hunt. Jev + Treg 👇

Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇

Athrix ☄️
@athrix_codes

My whole job hunt just cost me 3.7 cents 281 open roles, 12 scored against my profile by Jev, the pay, and who to write to at composio Job hunting cooked! Here it is:

Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇

Mashood K
@fromcodetocloud

Stop wasting hours jumping between job boards to find your next Devops role. I tried Treg's new Jobs API powered by Jev to find Devops, Platform Engineering, SRE and Cloud Infrastructure openings matching my skills and experience. It helps me discover relevant jobs without manually searching multiple platforms. PROMPT IN COMMENTS !

Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇

Lenny Enderle
@lenny_enderle

the job hunting trick nobody tells you: don't write to the recruiter, write to the person who owns the role gave my agent my website instead of a cv. treg pulled every open ai role remote in europe or lisbon, jev ranked 82 of them against my projects, and it checked what they actually pay then it found the team lead for my top 3, verified their emails and drafted a note linking the project that fits best. nothing sent without me, $0.38 total

Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇

xsmryze
@xsmryze

THIS IS AN ACTUAL F*CKING GOLDMINE Anthropic just shared one of the best tricks for Claude Code. The idea: run Sonnet, Haiku 5.5 and Opus 5.5 at once, each one doing only the job it's best at. Better results and way fewer tokens, because every task lands on the right model. You just launch Claude Code like this: claude --model sonnet --advisor opus here's the split: → Sonnet 5.5 runs the session, writes the code and applies the fixes → Haiku 5.5 swarms the repo in parallel: search, grep and diff audits → Opus 5.5 stays on call as advisor for the big calls and signs off on the diff → you stop burning Opus context on work that doesn't need it Opus only steps in at three moments: → before a plan locks → when the same error shows up twice → before calling it done Sonnet builds. Haiku swarms. Opus reviews. JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split the full tree > Sonnet 5.5 runs the main session and coordinates the work > search maps the repo on Haiku 5.5, read-only > grep finds call sites on Haiku 5.5 in its own context > audit checks the diff on Haiku 5.5 in a throwaway worktree > Opus 5.5 on call as advisor, reviews the synthesized diff before sign-off > JEV routes micro-forks in milliseconds paste the tree and this prompt into Claude Code ↓ "Reconfigure my Claude Code setup around this tree: 1. Run the main session on Sonnet 5.5 with Opus 5.5 as advisor. 2. Create three Haiku 5.5 subagents in .claude/agents: search (maps the repo), grep (finds call sites), audit (checks the diff). 3. Set isolation: worktree on audit so its checks never touch my working branch. 4. Add one rule to CLAUDE.md: delegate every repo search and test run to the Haiku 5.5 subagents to keep the main context small. 5. Route fast deterministic forks (tool choice, path selection, retries) through JEV. Show me every change as a diff first. No edits until I say go." ↳ save it, it'll come in handy.

rody
@0x_rody

this is insane... this pdf just broke the economics of Grok Bot teams. it shows how to put a Jev call on every handoff between your bots: which bot takes the task how hard it should try whether the action is safe whether the draft is good enough drop the pdf into your Grok Bot and let it rebuild its own team around it. it takes one evening, and your bots end up on a stack almost nobody else has:

Hanako
@hanakoxbt

ok this harness is f*cking insane Claude Fable 5.2 runs 1,000 coding agents on one job, and the harness decides whether it finishes the argument fits in one line: a thousand agents need briefs, not prompts a brief is one piece of work that edits files no other piece touches a builder returns a receipt a stranger checks it, a verifier that never saw the builder's reasoning Jev Engineering sits on top it routes every prompt to a seat, checks GOAL.md in about 100 ms and sends anything it's unsure about up deletes, pushes and deploys always route back to you so your job shrinks to two things: write the brief, approve the deploy the whole guide in one line: give every agent one job and a stranger to check it then read the article below ↓

Artyom Shimanski
@a_shimanski

i let Claude Code run my job hunt for $0.05 with it: ⤷ pulled remote AI roles posted this week ⤷ ranked each one against my resume with Jev ⤷ found the CTO and VP of tech to message ⤷ checked what the market really pays the part that hurt: remote Java devs get $121k median remote AI engineers get $171k same me, +41% 👇

Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇

Browse by category

Jev topics

Explore the latest discussions on X.

FAQ

About the Jev model

What is Jev?

Jev is the first System One model from TypeSafe AI, built for automation and software decisions. Rather than generating conversational text, it takes an input state and directly returns typed decisions with calibrated confidence that software can use. Its training method is called RLCD (Reinforcement Learning for Calibrated Decisions). The official website claims that Jev is 193.6 times faster and 444.6 times cheaper than conventional LLMs for automation tasks. These figures come from the official website and have not been independently verified by this site.

How does a System One model differ from chat models such as ChatGPT and Claude?

Chat models excel at generating long text and performing multistep reasoning. System One models focus on quick decisions: given some content and a question with a clearly defined output type, they return a choice, a score, or a true-or-false judgment, along with the probability of each outcome. They suit decisions that software makes repeatedly, such as classification, routing, scoring, moderation, and tool selection.

What is RLCD?

RLCD stands for Reinforcement Learning for Calibrated Decisions. It is the method TypeSafe AI uses to train Jev. The goal is to align a model's confidence with its actual accuracy. For example, judgments made with 90% confidence should be correct about nine times out of ten, allowing software to use confidence to decide whether to act automatically or hand the decision to a person.

What kinds of questions can Jev answer?

The official documentation defines three question primitives: Choice (selects one option from a set and returns each option's probability and confidence), Score (assigns a score against a rubric with 2 to 10 levels), and Noul (judges whether a statement is true and returns a probability from 0 to 1). A single request can ask multiple questions.

How do I get started with Jev?

Get an API key from the TypeSafe console, send a POST request to https://api.typesafe.ai/v1/systemone, and set the model name to jev-latest. See the official documentation for details. Official Python and JavaScript SDKs are also available. According to announcements on X, Jev has also been integrated with OpenRouter, Cloudflare AI Gateway, and Vercel AI Gateway.

How much does Jev cost?

As of September 2026, the official documentation lists input at $0.042 per million tokens, with free output and a limit of 64k tokens per request. Prices and limits may change; refer to the official documentation for the current terms.

About this site

What does this site do?

This site curates quality discussions in Chinese and English about the Jev model on X (Twitter), including open-source projects, integrations, tutorials, hands-on reviews, use cases, interviews, and news. It also tracks discussion activity by date. The site is maintained by the community and curated by @Web3wr. It is not affiliated with TypeSafe AI.

Where do the posts come from, and how often are they updated?

Posts are retrieved through twitterapi.io's advanced search API. The search includes posts mentioning both Jev (or System One or RLCD) and TypeSafe, excludes reposts, and also includes the latest posts from the official @typesafeai account. Popular picks refresh every hour, while daily statistics are collected in batches by the backend every 5 minutes.

What should I do if a category or ranking seems wrong?

Email hello@jev-x.com, or send feedback through Telegram or message @Web3wr on X. You can also recommend quality posts that we have missed.

Why do videos in posts sometimes fail to play?

Videos load directly from X's video servers. If your network cannot access X or X restricts playback, the card displays a notice and directs you to the original post on X to watch the video.

Ranking and statistics

How are posts ranked?

Daily picks and the 3-day hot list are ranked differently. Daily picks are ordered by Jev quality score from high to low, with newer posts first when scores are tied. The 3-day hot list only includes posts published in the last 3 days and is ordered purely by X engagement score from high to low, with newer posts first when scores are tied; the quality score is only used to filter out noise and does not affect the order. The X engagement score uses weights based on the recommendation ranking model X open-sourced in 2023: 0.5 per like, 1 per repost, and 13.5 per reply. How popularity is ranked → Picks rules

Why not rank posts directly by views?

Views reflect exposure, and large accounts get many views regardless of what they post. Replies also carry a high weight in X's model, but posts such as giveaways asking people to reply with "1" can inflate their scores through reply farming. So this site ignores views: daily picks use Jev to judge how informative the content itself is and are ordered directly by quality score; the 3-day hot list is ordered by X engagement score, but first uses the quality score to filter out noise such as giveaways and engagement farming (posts with a quality score below 50, a failing grade, are not shown).

How does Jev assign quality scores?

Each post is sent to Jev for classification and a Score question with 5 levels, then converted to a score from 0 to 100: noise (giveaways, engagement farming, or irrelevant content), pure emotion, an opinion, specific information (news, integrations, demos, or measured results), and worth saving (original insights, tutorials, open-source tools, or data). Jev also sees the types of links attached to the post, such as articles, GitHub, and videos, as well as engagement data such as the bookmark rate. Exaggerated promotion in all caps does not earn extra points. Post quality → Picks rules

Do quality scores change over time?

Each post receives a quality score only once. The X engagement score is recalculated from the latest numbers on each refresh, and the 3-day hot list is ordered by it, so a post can still rise in the rankings if it becomes popular later.

How are the post, view, reply, repost, and like totals at the top calculated?

Since Jev's release on September 15, 2026, the site has counted qualifying discussions about Jev on X each day: related posts with at least 10 likes, excluding reposts. The numbers are snapshots taken when posts are collected; view counts can continue to grow afterward. Long-tail posts with 0 likes are not included. Dates and times follow your time zone, matching X: each post is grouped under its publication date in your local time. Statistics use your local date from October 2, 2026; earlier statistics only exist as UTC-day totals and are kept as originally recorded.

Content categories and picks

What content categories are available?

Official (posts by @typesafeai), Integrations (platforms or frameworks integrating Jev), Open source (GitHub projects, CLIs, plugins, Skills, and open-source recreations), Competitors (comparisons with ChatGPT, Claude, Codex, and others), Tutorials (getting started, study notes, and explanations of concepts), Use cases (applications and demos built with Jev), Reviews (measured speed, cost, and accuracy), Interviews (podcasts, interviews, and talks), News (releases, funding, and user growth), and Opinions (other views).

How are categories assigned?

Jev assigns categories through a Choice question. Each category includes a definition of what belongs and what does not, along with real examples. Jev's result is used when its confidence is at least 50%; below 50%, keyword rules provide a fallback. Hover over a card's category label to see the source and confidence.

How are daily picks selected?

Each day, classified qualifying posts are ordered by Jev quality score from high to low (newer posts first when scores are tied), and the top 50 are selected. Posts with a quality score below 50 (a failing grade) are not shown. The current day's data updates every 8 hours, is collected again after the day ends, and is finalized after another 24 hours.

Share preview · Jev-X

Image preview