Jev for data extraction — a daily feed

jev に文字起こしテキストと show notes 渡して、リンクのトピックが文字起こしの何分何秒にでてくるかを判定。3時間のエピソードで 0.5s, $0.003 でできる https://t.co/KdPA1MYQEX

luccafort さんのプロフィールを 8,000 文字用意して jev にデータを入れました。 AI で作ったので、レビューしてないです。 #byebye_lucca https://t.co/DaK7mtsp3a

JEV is INSANE It read 6,030 ChatGPT answers to buyer questions about 1,005 YC companies and checked every company name in every answer. 37,636 verdicts. 9.1M tokens. 40 seconds. $0.38. Grading the same 6,030 answers with Claude Sonnet cost me $36 last month. 71% of these startups were never named once, even when the buyer asked ChatGPT for exactly what they build. Scan your site and see if AI ment

Pub/Sub AI Enrichment with Gemini and Jev https://t.co/YdSqFmaUsR

Ask an LLM for JSON and it types it out char-by-char while you parse and pray. Jev doesn't: hand it text + a schema, get a typed value in one pass — 0% malformed, ~0.4s, 1/76 the cost. No text, no rationale, bounded only. → https://t.co/2OUi9ofr1Z #LLM #StructuredOutput #AI https://t.co/K1kDt3rZLU

使用 #Jev 做了一个填写外链信息的演示。 眨眼的功夫就找到入口，并准确的把已知信息填写完了。 准备把之前的Submit Agent改成使用JEV模式。 关注github，在评论区 https://t.co/Ffwex3kxqS

I’m surprised Jev for tabular data isn’t a bigger deal. Every ai spreadsheet company should be racing to rebuild their product ground up with Jev. When we were building people search last year we spent months optimizing the search algorithm. Here is Jev processing 100k rows for $2.50 in under 60s, no optimizations whatsoever. Public demo at https://t.co/6kK0XAURtj try it out with your own csv. A f

受启发也搞了一个，用 Jev 识别剪帖板信息实现表单自动填充 https://t.co/pvYnvmHBGA

有報からtypesafeaiでTOPIX1000銘柄全部で定性スコアを多次元抽出して組んだL/SポートのFF3残差の時系列推移。これ作るのにFableにお願いして１時間+typesafe API代金５ドルだぜ、笑っちゃうね。 https://t.co/ZXD5CUPJcp

話題のJevを使って個人情報を一気に入力する機能を作ってみた。もう個人情報をちまちま入れることから解放された。ガチでAIの進化が凄すぎる。 遊んでる暇はない、AIに適応した人としていない人の差が、必ず顕著に現れる日がくる。世界が変わる。しかし、AIのしすぎで肩と腕がいたい。時間が溶ける。 https://t.co/oU28nsk8Cv

Jev is cool, but this is OWL: 24 free api based web scrapping agent running on a ~25mb embed model inside of indexedDB. EDGE AI in a minimal format: Robust planning/git cloning/price matching using only free API: https://t.co/81Bvg6ICF5

turned the Constitution into a knowledge graph in 4s, the Odyssey in 45s. - per sentence, deterministically generates possible subject, predicates and objects. - used jev to sift through it all. result is very fast and cheap triple creation. https://t.co/vfFawqABOZ https://t.co/pn0tphOAvI

never forget your favorite cup of coffee super proud of my submission for the all gas hackathon we've got: - @convex components galore (static hosting, auth v2, agent, aggregate, workpool, rate-limiter) - @firecrawl to scrape coffee roaster product pages, paired with a little jev action to classify the data - @openai's 5.6-luna in a convex agent to recommend your next bag of coffee by comparing yo

Curating clinical variables from free-text notes is tedious. General LLMs can help, but processing thousands of notes can be slow and costly. Inspired by Jev and the open-source community, we’re releasing MedJev to turn clinical notes into structured fields on consumer GPUs. A 0.8B model + 43 MB LoRA adapter, trained to extract 11 predefined clinical variables. On our benchmark of 2,895 held-out n

can @typesafeai Jev count the r's in strawberry? no. 47% says 3, 47% says 2. a coin flip, same as every LLM. On 70% on 168 test words, it undercounts doubled letters like everyone then I gave it the letters instead of the word: ["s","t","r","a","w","b","e","r","r","y"] 168/168. same model, same question, 260ms @CompleteSkeptic is this expected?

I built a tool powered by Jev that extracts any products features and business logic into a markdown repo Enter a product URL and you can copy features, ICP, messaging, pricing strategy, etc straight to a md file https://t.co/mS8CoBuk1x

design experiment with jev + astra + gpt voice + maplibre + apple photos + gps metadata i connected all of this just so i could name places over spongebob music and fly around a map with my vacation photos 💀 https://t.co/jo0FVDuiRq

@marcus_lowe @typesafeai I built jevfill, a chrome extension that auto fills from notes https://t.co/gfTejUtz91 https://t.co/QmlQD6AIKY

@typesafeai @typesafeai I made a module for Magento 2 (most famous e-commerce platform) that uses Jev to analyse order details (like fraud, etc), customers, products, marketing, abandoned carts.. etc https://t.co/4LVuGu8dqv

JEV invoice arbitrage is getting out of hand... Bookkeepers charge ~$0.50 per invoice for manual entry. JEV reads 1,000 invoices for $0.40. That's a 99.92% margin. Vendor, invoice no, dates, VAT, total, GL account. All 8 fields, ~2 seconds each, 6 workers in parallel Thanks for the opportunities, Typesafe! Accountants, I'm sorry

Inspired by Jev Ultrafast - JevBro(wser) powerd by @typesafeai jevbro is a CLI for real work: sessions on disk, resume when a field is missing, extract that refuses invented fields, guards in code, a coding-agent skill, local text model if you want it. Demo: rubber duck → Apollo 11 on live Wikipedia. 6 steps. $0.006. Title + URL extracted from the page. https://t.co/sVRBHfR7OZ

. https://t.co/3WXU3WFDzF JEV integration is now live and working, enabling smarter decisions on which keys to output on-screen for keyboards with broken keys. https://t.co/g6GMed8Gfy

「AIに全部見て」をやめる。 重複や空欄はルール、意味の食い違いはJEV、原因調査はCodex / Claude。候補と確定を分けて記録する実践スキルキットをBrainで公開しました。 1,980円 https://t.co/6NqucD6tTy https://t.co/CGHD3ci5eF

Hopefully not just another weekend Jev project. https://t.co/mdCBEWIZ4A Used Jev to extract billing codes (aka ICD10) from medical claims. 88% accuracy to 3rd character of precision with claims that have enough information. Medical coding is key for correct healthcare billing

Gliner is well worth looking at. I've used it in the past, but looking forward to seeing what is new. You can see some good tutorials here: https://t.co/7bzSdbg4mD Also suggest looking at DSPy. You can see the classification tutorial for email here: https://t.co/yiuPogAeWq I played with Jev quickly for a simple extractor for populating a JSON-LD schema,. Think is would be fun to try the same appro

話題のJevで何作ろうか迷って、 2ヶ月前に「Gemini Nano」で実験して頓挫した 打ち間違い修正ツールを作ってみた 「メール打ってたら英数入力やん…😭」ってことよくあると思うんですけど、それを日本語に直してくれる。 それだけ（笑） あと誤字もまぁまぁいける。 自分はほぼ音声入力なので正直使わないそうですが、 音声入力が気まずい職場とかならちと便利かなと。 そしてこういうのはUXが命ですね。候補チップが邪魔なときある😂

This one is a little more out there and not really ready yet, but because @rsg asked if it was possible, I had to try it. This attempts to use Jev to extract a "reader mode" for content. And it... sort of works? It butchers many examples, but so do many existing code-based attempts. This has given us ideas for more targeted applications of content extraction.

Free Jev access on @vercel AI Gateway is awesome! I made an OSS library & x402 API to do text extraction with Jev. Who said Jev can't output text 😂, all you need is to return indexes to the input and slice the text! 100% @typesafeai Jev w/ no LLM! Demo + repo below 👇 https://t.co/mulwMs4E1U

Jev + Mailient = forgotten promises solved ✅ 177 questions/meeting in ~1.7s for $0.0006 🤯 > paste meeting transcript > reads it > splits what you owe vs. them > ignores unowned “someone should…” SO: paste, see what you promised, chase it before it dies. Free, no signup ↓ https://t.co/v1Pbwep6fW

LLM overhead kills agent loop economics. Routing decisions, relevance scoring, next-worker selection. Each call costs $0.03. In a tight loop that's your entire budget. I built DeepScrape with the same principle in mind. One LLM call to derive a CSS selector. Every subsequent extraction is deterministic. No more paying per-extraction for structured web data. Same logic Jev applies to routing decisi

Using @GROK to make json snippets for @typesafeai https://t.co/XJhi979vSu

【my_web_app 画面操作手順】 ①トップ画面右上のログインボタンからログイン ②例として「Googleでログイン」を選択 ③ログイン後、人気の機能「Asset Management」を選択 ④資産管理画面へ遷移 ⑤支出メモ入力でJevが支出カテゴリをミリ秒即時分類！ https://t.co/j3v61tiVpe https://t.co/1jm8zOJCYi

Grounding DINO gives a box, SAM 3 gives a mask. The mask fused with RealSense depth tells us where solid material is and how wide, so the grasp goes to a line across the part with room for both fingers. For the chrome parts depth is useless (the camera sees the ceiling in it), so we fit the part's STL silhouette to the SAM 3 mask instead, table contact fixes the height. Then https://t.co/URMcA9opc

Jev × BI探索「セマンティック マスタマッチング」 データを取り込む際の項目マッチング的なやつで、文字列での機械的マッチができないものは人間がマッピング作業をする、というペインに当てる Jevの意味としては: ・文字列でマッチできないものも意味的にマッチングできる（自身がないものは判断を促す） ・間違っていた場合は選択し直すが、別の関連性が高い候補順に出るので訂正が楽

First real use case with Jev got up and running last night. Used it with GLiNER 2.5 base to help dramatically speed up entity / relationship extraction. GLiNER by itself for Graph RAG over a huge document corpus is fairly unreliable. Using Jev to help determine what it got right and then send everything else to my local 35B model. ~25x performance boost.

Update: 1- Introduced gpt-oss-120B for extracting Plain Requirement from JD. Now keyword extraction from JD is more precise. #jev Try it out : https://t.co/Rnkm6bfNOJ