plog retweeted
Someone just built something WILD for Claude Code users.
A self-learning system that automatically creates, updates, and organizes its own skills.
Instead of repeatedly teaching Claude the same workflows, you let it learn from real sessions.
This could change how people use coding agents.
It's called Autoharness.
No manual SKILL .md writing. No pile of duplicate skills. No background daemon eating your machine.
It watches your real Claude Code sessions, pulls out the lessons, and turns them into skills Claude loads next time.
Here's what this thing can do:
→ Writes new skills from the work you already did
→ Merges near-duplicates instead of stacking them
→ Rewrites old skills when you correct it
→ Archives skills that stop getting used
→ Opens every session with an index of what it learned
→ Type /learn to save a lesson on the spot
→ Never touches skills you wrote or installed yourself
Install it inside Claude Code:
/plugin marketplace add tigerless-labs/autoharness
/plugin install autoharness@autoharness
github.com/tigerless-labs/au…
plog retweeted
This is what AI is made for. Murtaugh is too old for this 💩 and Riggs has to find a new partner. I absolutely love how they brought in every character from that era and even combined two different characters for one actor in a couple different scenes. 🤣🤣🤣💥🔫
plog retweeted
Every Mac on macOS 27 now has a free LLM built into the Terminal.
No account. No API key. No internet. No token bill.
Setup (1 minute):
1. sudo fm license (enter your Mac password, one time)
2. fm available (checks the model is downloaded)
3. fm respond “your question”
Works on any Apple Silicon Mac, M1 included.
- Pipe files: cat notes.txt | fm respond -i “summarize this”
- Get JSON back: build a schema with fm schema object, pass it with --schema
- Local API: fm serve gives you an OpenAI-compatible endpoint on localhost
- Resume chats: sessions save to ~/.fm/sessions/
The limit is 4,096 tokens per session, and your prompt, instructions, and the answer all share it.
So skip code and math. Use it for summarizing, extracting, classifying, and rewriting.
Free, private, offline, and already on your Mac.
plog retweeted
🚨 CHINA JUST OPEN SOURCED AN OCR THAT EATS 100 PAGE PDFS IN ONE PASS
It’s called Unlimited OCR. Only 3B params. Runs locally.
Every other OCR tool chops your doc into pages and loses the thread. This one reads the whole thing in one continuous pass.
→ One shot full context parsing (32K context window)
→ Multilingual out of the box
→ 93% on the standard parsing benchmark (6 points over baseline)
→ Under 0.11 error rate past 40 pages
→ Runs 100% locally on your own hardware
→ Works with Transformers, vLLM, SGLang, Docker, Ollama, llama.cpp
Traditional cloud OCR (Textract, Google Vision, Azure Doc Intelligence) costs $1.50 to $15 per 1,000 pages.
This runs on your machine. For free. Forever.
Baidu built it explicitly to push DeepSeek OCR one step further. Already at 1.9M downloads on Hugging Face and most people have no idea it exists yet.
100% open source.
plog retweeted
I don't understand why everyone isn't doing this yet. Anthropic released a guide for Claude that makes work much more efficient while burning a fraction of the tokens
Main Idea: Agent Teams in Claude Code
Instead of burning the expensive Opus 5.5 context on every line of code, tasks are delegated across an autonomous swarm:
Architect (Opus 5.5, high effort): Plans the structure and handles final PR merges
Developers (Sonnet 5.5, medium effort): Write code inside isolated git worktree environments (UI/UX, backend)
Adversary / Critic (Fable 5.1): Steps in at contract boundaries, recurring test failures, and pre-PR audits to catch vulnerabilities
Script Layer (Jev): Handles mechanical tasks (opening files, CLI invocations) in 16ms without involving LLMs
Launch the team via terminal:
claude --teammates ux,backend,adversary
Configure my Claude Code workspace for autonomous agent teams:
In ~/.claude/teams, generate missing roles: set workers to Sonnet (medium effort) in separate git worktree branches, and the adversary to Fable
In ~/.claude/settings.json, set CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 and lock the main session to effortLevel: high
Initialize tasks.json with file-level locking for direct peer coordination
In CLAUDE.md, add a rule requiring an adversary review before opening a PR.
Show configuration diffs before applying edits
> code.claude.com/docs/en/advi…
plog retweeted
Another insane Jev use case!
Traditional database filters need precise, predefined conditions. But many questions are semantic:
- Is this article mainly about software engineering?
- Which topic best describes it?
- How technically deep does it appear?
These usually require moving rows into application code, invoking a model, parsing its output, and writing the result back.
pg-jev is an open-source Postgres extension that exposes Jev through SQL functions.
- jev() works inside WHERE
- jev_prob() returns a probability
- jev_choice() selects a label
- jev_score() ranks rows across ordered levels.
It needs no vector column or embedding index.
Under the hood, pg-jev batches rows, sends them to Jev, and returns typed answers that SQL can filter, sort, group, and combine with exact predicates.
The recording below runs it against real Hacker News stories stored in Postgres.
It first shows the rows, then asks Jev to find software-engineering stories, classify them by topic, and rank them by technical depth. The final query reports requests, tokens, cost, and cache hits from the database session.
GitHub repo: github.com/realZachi/pg-jev
(don't forget to star it ⭐)
Postgres keeps the data and controls the query, while Jev handles the part SQL cannot express as a deterministic condition.
If you want to understand what Jev is doing under the hood, I also wrote a hands-on guide to building a Jev-style model with open models, entirely locally.
Read it below.
plog retweeted
Nothing good can come of this. AI can now generate video in real-time. Explainer below.
It just feels like this will only be put to bad uses...
What say you? Comments welcome.
plog retweeted
i'm still not sure people are fully getting how good opus 5.5
i gave my claude agent fig access to the runway MCP and told it to make a 'high-end netflix style documentary about superintelligence for normies' and...
it came back with this
plog retweeted
TyperSafe Founder, Diogo Almeida just gave the best 2-hour breakdown on how to actually master Jev, from 0% to 100%:
• 10% → 21:43 - building an eval for your own workflow with JEV
• 30% → 55:49 - choice, score, noul - new types of code mapping with JEV
• 55% → 1:12:36 - cascade: confident answers ship, uncertain ones go to a bigger model
• 80% → 1:37:12 - JEV use cases: dark data, real-time calls, verifying every LLM in your stack
• 100% → 2:11:50 - building coding agents with JEV
this is worth more than another $500 course on building agents with Claude
watch it today - then read the full Jev playbook below
plog retweeted
Jev has been exploding in popularity recently.
If you already have access to the Jev API but aren’t sure how to start experimenting with it, just copy this checklist:
1. agent-desktop
Desktop automation. Read the system's accessibility tree, judge which button, menu, or input field to click next. github.com/lahfir/agent-desk…
2. typesafe-mario
Have Jev play Super Mario. No screenshots—just read the structured state in the emulator's RAM, then decide to run, jump, or dodge. github.com/fhshaik/typesafe-…
3. jev-drone
Use Jev to control a drone. The underlying flight control still handles stability and safety; Jev just does higher-level judgments like climbing, braking, and navigating obstacles. github.com/RomanSlack/jev-dr…
4. OneVOneJev
1v1 FPS in the browser. Every decision tick, judge movement, view angle, aiming, firing, and jumping. github.com/emrickgarrett/One…
5. jev-trader
High-frequency market making on Monad testnet. Jev judges the next buy or sell based on spreads and trade direction, with model latency around 81ms. github.com/jarrodwatts/jev-t…
6. Prism
Doesn't directly have Jev place orders. It judges states like toxic flow, market pressure, mean reversion, etc., then hands off to the original strategy. github.com/irfndi/prism-liqu…
7. neo4jev
Stuff Jev into a knowledge graph. At each node, judge the most worthwhile edge to take next, then follow it all the way. github.com/jexp/neo4jev
8. jev-curate
Use Jev to screen training data. For JSONL / Parquet, first judge quality, relevance, and risk, then decide which ones go into the next training round. github.com/AkashPriyadarshii…
9. Canny
Prevents Coding Agents from stubbornly claiming they're done. Look at tool outputs, code diffs, and test results, then judge if the completion claim is reliable. github.com/qkal/Canny
10. killmyidea
Input a startup idea, and Jev scores it from multiple angles, finally giving you KILL, FIX, or SHIP. github.com/monteduro/killmyi…
Copy these complete Jev blueprints - then read full Jev setup below ↓ ↓
plog retweeted
We need to talk about Jev
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
plog retweeted
The DESIGN resources you need, save this 👇
- reelfolio.io Showcase your work in style
- inspora.design Discover the best design inspiration
- backgrounds.supply — Beautiful backgrounds
- fontshare.com — Free, quality fonts for designers
- hugeicons.com — Massive icon library for designers
- remove.bg — Remove image backgrounds instantly
bookmark these tools for your next project.🔖
plog retweeted
I built a Photoshop-like image editor. It’s called Compositor.
robbietilton.com/compositor
I originally built it for myself to get off my Adobe subscription, but decided to release it for free and make it open source. It has all the essential tools I need for compositing, with none of the BS.
I know this workflow is kind of archaic in 2026, but I’m still using it until AI can actually get pixel-perfect on some of the details.
The entire app is 12MB. Photoshop is 6,455MB on my machine.
plog retweeted
👋 ADIOS Arquitectos.
👋 ADIOS AutoCAD.
👋 ADIOS licencias de $10.000/año.
Existe un editor 3D de edificios GRATIS.
Se llama Pascalorg, y tiene +24.000 estrellas.
Te dejo el link abajo: ↓
plog retweeted
Artemis Mission Route in 3D
- This animation visualizes the Artemis mission trajectory in a dynamic 3D perspective, showing how the spacecraft travels through the Earth–Moon system while all celestial bodies are in motion. Instead of a static path, the Sun, Earth, and Moon move simultaneously, revealing the true complexity of orbital mechanics. The result highlights how the Artemis route is not a simple curve, but a constantly shifting trajectory shaped by gravity and motion. This view provides a clearer understanding of how modern space missions navigate through space in real time. Right now, Artemis is on its return path to Earth and is expected to arrive back soon as it completes its mission. The sizes and distances of the Sun, Earth, and Moon are not to scale and are adjusted for visual purposes.