←
CodeWorks3,342 posts
CodeWorks avatar

CodeWorks

@CodeWorksParis

💫 L’ESN solidaire au service d’un code utile, fiable et maintenable. #craft #web #data #cloud

◷ Joined December 2008▧ 129 Photos and videos
CodeWorks avatar
CodeWorks@CodeWorksParis·7 May 2024

@CodeWorksParis est sponsor de @ncraftsConf ! 3 de nos CodeWorkers auront la chance de participer aux 2 jours de conférence. 🗓️Et le 16 au soir, venez assister au meetup @WomenTechmakers. Bravo aux @DuchesseFR @LadiesCodeParis pour la co-organisation ! 💫

📢 Remember to save the date for the next @WomenTechmakers Paris meetup: May 16th: IWD24 Paris: Crafts-Women Techmakers & Friends (@duchessfr, @LadiesCodeParis) @ncraftsConf Conferences [Mandatory Registration]: lu.ma/joisxg95

Post attachment
Polash Khan avatar
⟳ CodeWorks retweeted
Polash Khan@polash_ai·Nov 8

It worked. Gave Claude $20. Left it coding overnight. Woke up to: • finished script • +$95 from first 5 trades Didn’t touch anything. Then I found this wallet: 0xb27b...b82 • $384,858 profit • 13,070 predictions • $18.5K biggest win • Started March 2026 Almost no one is watching it. While building my script, I copytraded it to test the math. Started with $80. → $4,218 in a few weeks Every position is identical in structure. Not trading. One equation. Running while you sleep. Examples: $17,839 → $36,318 (+103%) $3,112 → $15,011 (+382%) $10,676 → $22,178 (+107%) Here’s the exact framework I gave Claude: 1. Temporal bias Find BTC time windows with >65% edge. Ignore everything else. 2. Kelly sizing Size based on confidence. Never fixed bets. f* = (p × b − q) / b 3. Volatility filter Skip high volatility. Trade only clean conditions. 4. Bayesian updates Update probabilities after each trade. The system improves over time. Claude built everything while I slept. Funded it with $20. First trade fired minutes later. Copytrade my finds → @polash_ai … You’re seeing this early. That window doesn’t stay open. You only need Claude + Device + 1 hour/day. Giving This Free for 24 hours. To get it: 1. Comment the word ' DM ' 2. Like and Retweet this post 3. Follow me @polash_ai

0:31
Kshitij Mishra | AI & Tech avatar
⟳ CodeWorks retweeted
Kshitij Mishra | AI & Tech@DAIEvolutionHub·Sep 19

Anthropic senior engineer just released a 1-hour course on building a team of agents with loops & graphs: • 00:27 - introduction to CLAUDE.md & Plan mode • 11:24 - building "skills" & "hooks" for Claude agents • 37:02 - building agents & subagents with Claude • 52:47 - self-improving loops & graphs for Claude agents this 1-hour watch will replace a $500 agentic engineering course watch today, then read how to build a team of self-improving agents that work together

Post attachment
rari🈀 avatar
⟳ CodeWorks retweeted
rari🈀@0xwhrrari·Sep 19

Google Brain founder Andrew Ng: "Prompting will be dead in 6 months Agent harnesses built with loops and graphs will replace it" Agent → Harness → Feedback → Loops → Graphs → Self-Improving Systems In this 1-hour Stanford lecture, he explains what the best engineers are building instead and how you can start today Prompt → Run → Verify → Improve The first 20 minutes teach what most $1,500 courses try to sell you For free Bookmark and watch it today Then read the guide below on building an agent harness that improves itself

1:40:11

rari🈀 @0xwhrrari Aug 29 Article Harness Engineering: How to Build AI Agents That Don't Fall Apart Most people respond to a failing agent by changing the prompt Then they change the model Then they add a larger context window The agent still forgets decisions It still uses the wrong tool It still

Mahax avatar
⟳ CodeWorks retweeted
Mahax@Mahaximus_·Sep 18

Google just dropped the best 1-hour course on Graph Engineering: from one agent to a full 24/7 system 00:00 - What Graphs are 09:16 - Build an agent 21:15 - Graph engineering explained 41:03 - Graph engineering practice 52:21 - Self improving Graphs Free, the best thing on Graph engineering I've come across Watch it, then read the full guide on agents and graphs below.

1:00:22

Mahax @Mahaximus_ Aug 5 Article Agents, Loops, Graphs. Everything You Need to Know in One Place. Most people use AI at one speed: ask, read, fix, ask again. It works. It's also the slowest way to do it. There is a faster setup. It has three parts: an agent that plans and acts without being walked

sergio avatar
⟳ CodeWorks retweeted
sergio@sergiorocaa·Sep 19

Este ingeniero de Anthropic explica la manera correcta de construir agentes de IA en 14 minutos. La mayoría de los desarrolladores pasan meses haciéndolo de manera equivocada. Guarda esto antes de escribir otra línea de código de agente.

14:46
Mahax avatar
⟳ CodeWorks retweeted
Mahax@Mahaximus_·Sep 19

instead of watching 2 hours of Netflix tonight, watch this Stanford lecture given by Anthropic engineers it's the clearest explanation I've seen of how AI agents actually work useful whether you've never touched AI in your life or have been building with it every day i took the key ideas and turned them into a practical guide, with ready-to-copy prompts watch it, then read the guide below on the easiest way to build your own AI agent team that improves itself over time

2:12:36

Mahax @Mahaximus_ Aug 5 Article Agents, Loops, Graphs. Everything You Need to Know in One Place. Most people use AI at one speed: ask, read, fix, ask again. It works. It's also the slowest way to do it. There is a faster setup. It has three parts: an agent that plans and acts without being walked

Ajit kumar avatar
⟳ CodeWorks retweeted
Ajit kumar@ajitcodes·Sep 19

Anthropic engineer: "99% of people use Claude Code like Google, and only 1% are running swarms of self-learning Claude agents I'm running 100+ agents in a loop. I have Chief agent, PM agents - they manage the whole team" in a 30-minute workshop, an Anthropic engineer revealed how to get max value from Claude Code at min. cost this is worth more than another $500 vibe-coding course watch today, then read how to build self-improving agentic systems with Fable in the article below

26:36
elvis avatar
⟳ CodeWorks retweeted
elvis@omarsar0·Sep 18

Build your own harness, folks. This is absolute banger paper from NVIDIA on self-evolving agent harnesses. (bookmark it) They introduce SoL-Pi which cuts token traffic by nearly half. And it matches its baseline harness on GPT-5.6 Sol and Opus 5. More details below: Instead of tuning a harness by hand, they run auto-research loops at the harness layer across many repository-derived and verifier-driven environments, keeping only the mechanisms that survive selection. Four mechanisms survived: > Action Fusion changes how actions execute > Online Context Compact handles compaction during a run > ObservationPack reshapes observation handling > Evidence-Preserving Reducer covers delegated reading On the 51-task EdgeBench evaluation, the savings translate to about a third off API cost. In dollars that is an estimated $8.75 to $13.50 per hour against native Codex and Claude Code harnesses, and $4.36 to $5.71 against the baseline harness. Because the search runs across many environments rather than one, the retained mechanisms keep working outside the setting that produced them. Code is on GitHub under NVlabs. Paper: arxiv.org/abs/2609.20519 Chat with Paper: academy.dair.ai/papers/sol-p…

Post attachment
南晚Nanwan avatar
⟳ CodeWorks retweeted
南晚Nanwan@snwiki238337·Sep 19

一个系统化的 Agent 工程学习路线图手册:Agentic Engineering Handbook。 它把散落在 OpenAI 博客、Anthropic 工程文、SDK 文档、cookbook、论文里的 179 个精选资源,整成一份从"手写 Agent Loop"到"生产落地/eval/安全"的结构化学习路径。​ 学习路径分 7 阶段(Phase 0–6)​: Phase 0:从零手写 Agent Loop(基于 shareAI-lab/mini-claude-code,给了 v0–v4 可跑 Python 代码:bash agent → 模型当 agent → 结构化规划 → 子代理 → skills) Phase 1:Agent 基础(该不该建 agent 的 4 问检查表、Building Effective Agents 等) Phase 2:MCP & 工具生态 Phase 3:Context / Memory / Skills(含 Agent Skills 规范、上下文工程) Phase 4:Harness & 长程 Agent(mini coding harness 练习) Phase 5:Coding / Workspace Agents(Codex / Claude Code 风格工作流) Phase 6:Evals / Safety / Production(smoke/macro eval 套件练习) 仓库:github.com/keyuchen21/agenti…

Alex Volkov avatar
⟳ CodeWorks retweeted
Alex Volkov@altryne·Sep 18

This is actually insane. This uses @typesafeai Jev model, as a plugin in Claude to review all the un-nesseasary tool calls, and it takes 1s to run! Like, literally, 1 second to take my Claude session from nearly 1M to ... 86K tokens! 😮 Ask your claude to install it and be amazed Use this prompt ``` Install, and configure : github.com/tamaratran/fast-j… ```

Post attachmentPost attachment

found the perfect use case for @typesafeai Jev: instant compaction in 2026, why is compaction still a summarization prompt? Jev can make it instant by scoring every tool call and dropping what’s irrelevant

0:05
codila avatar
⟳ CodeWorks retweeted
codila@0xCodila·Sep 18

Jev is "Internet" moment for AI: up to 193x faster and 444x cheaper in tests with Claude Fable 5.1 and GPT-6 Astra Whar is Jev, how to use and unlock its real 100х advantage in my 10-page research: step 1 → meet Jev: LLM writes, agents act, Jev chooses the next move - split intelligence from execution step 2 → turn every agent fork into three primitives: Choice selects one route, Score measures a defined scale, Noul returns the probability of yes step 3 → build before getting access: use TypeSafe’s official adapter with OpenAI, Anthropic or xAI, then swap in Jev without rebuilding the graph step 4 → setup first Jev: one state, three parallel decisions, risk-based thresholds and a real queue your agents can execute step 5 → batch decisions instead of serializing them: 13 questions in one call ran 10x faster and 12.2x cheaper than 13 sequential calls step 6 → place Jev at every bounded fork: choose the agent, model, tool, browser action or human escalation, then read fresh state step 7 → benchmark the entire loop: Browser Use hit Google Flights in 7.1s, Every ran 777 checks in under 0.7s, Mobile Jev completed 9 actions in 21s step 8 → rank wide, read narrow: Jev cut wrong Hermes skill loads from 16.8% to 7.3% and pushed legal Top-10 retrieval from 38% to 62% step 9 → steal a system, not a prompt: Chief of Staff, model router, inbox firewall, research feed, browser controller and safety gate all use State → Questions → Action → Verify step 10 → keep Jev out of math, writing and irreversible execution: code computes, LLMs create, Jev decides, fresh state proves the result the result: one slow, expensive agent becomes an always-on decision machine that routes, scores and escalates in milliseconds Copy the complete 10-page Jev blueprint - then read full 10-step roadmap below ↓

Post attachment

Jev is the "Internet" moment for the AI industry It tells your agents and LLMs what to do next, in milliseconds and at almost zero cost If you set it up correctly, you will have the AI engineer’s stack for 2028 In this article, I show you how

code w/Guillaume 💻 avatar
⟳ CodeWorks retweeted
code w/Guillaume 💻@guillaume_rygn·Sep 19

Ce gars à regroupé tous les projets avec JEV qu'il a vus passer et les à rassemblés dans un seul site ! Pas mal pour voir les use cases et prendre un peu d inspiration ! Le site c est jevable.com 👇

Post attachment

Looking for inspiration on what you are able to do with Jev (by @typefaceai) this weekend? Built a website called Jevable (jevable.com) which showcases all the fun demos on X - filterable by a few categories. Feel free to add your own project by using the + button!

0:20
Morty avatar
⟳ CodeWorks retweeted
Morty@0xMortyx·Sep 19

Jev doesn’t make the LLM smarter. it makes 193x faster. 444x cheaperand testable Jev could be the missing decision layer between LLM reasoning and real-world agent execution. without a decision layer, you usually solve that with: more prompting → more context → more retries → more tokens Jev makes the fork explicit. for every state: Choice → which route wins? Score → how good is each option? Noul → how likely is the condition true? then execution stays outside the model. full Jev Engineering breakdown below ↓

Post attachment

Jev could become the control layer AI agents have been missing. Instead of spending 5–20 seconds and expensive LLM calls deciding every next step, it can route actions in milliseconds at near-zero cost. In this article, I break down how

Kyrox avatar
⟳ CodeWorks retweeted
Kyrox@kyr0stack·Sep 19

Google Brain founder, Andrew Ng: "Prompting will die in 6 months. Loops and graphs are what's replacing it." In 99 minutes he shows how to build agents that plan, execute and improve without you Prompts → Agents → Loops → Graphs skip a layer and it comes back as a failure you blame on the model by the time you find it the week is already gone that is the whole difference between using AI and having AI work for you watch it today, then save the full guide on loops and graphs below ↓

1:39:14

Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a

Ichigo avatar
⟳ CodeWorks retweeted
Ichigo@iiiichigo_chan·Sep 19

Google engineer: “Build a harness that runs self-improving agentic workflows for you 24/7 with loops and graphs. This is the new job! At Google, 90% of engineers already run dozens of agentic workflows through their harnesses. This is what the new era of engineering looks like” In this 1-hour conversation, a Google engineer breaks down what the future of AI engineering will look like Easily worth more than 10 paid agentic courses Watch it today, then read the article below

1:32:55

Ichigo @iiiichigo_chan Aug 29 Article Harness Engineering: Build a Reliable AI Agent in 6 Layers A better prompt can improve one answer. A better harness improves every run. If your agent can reason but still forgets constraints, chooses the wrong tool, skips verification, or loops until the