Jefferson Andres Espejo Goez

@Chicharrones_1 @Querpa @debittoinjapan @Eurolazaro Lástimosamente eso que dices no creo que aplique cuando utilizar una laptop corporativa. He trabajado en empresas que tienen esas practicas y están blindados para hacerlo porque al ser equipo de ellos, verán que software le instalan (nada de cámara obviamente)

@Querpa @debittoinjapan @Eurolazaro En un despido improcedente, procedente o de la forma que sea, un juez va a darle la razón al trabajador, básicamente es el derecho a la intimidad digital, RGPD, LOPDGDD, artículo 18 constitución, art20.3 estatuto trabajadores, es más ni una empresa da ese ejemplo para despedir.

El ingenio de algunos para que el Teams no se ponga en amarillo ausente

@kenwuuuu github.com/agentscope-ai/…

Both shipped for one reason: to help juniors and seniors actually use AI without burning out or burning tokens. gentle-ai: you ask, it resolves. Correct workflows, token optimization and resource best practices, all automated. Agent-agnostic, works with whatever you use. engram: MCP memory server with a queryable registry. Context survives across agents, and it searches the registry before re-reading code (massive token savings). Hundreds of hours of open source grind. What made it worth it: messages from people saying these tools saved their job or helped them land one, and companies running both in production. Built primarily with GPT 5.5 + different models for each SDD phase. github.com/Gentleman-Prog… github.com/Gentleman-Prog…

HARNESS ENGINEERING IS ABOUT TO CHANGE HOW YOU USE AI AGENTS Anthropic ran a controlled experiment. same model, same prompt, opus 4.5 no harness: $9 spent, 20 minutes, unusable output full harness: $200 spent, 6 hours, a game you could actually play the model didn't change... the environment around it did that environment has a name... it's called a harness and most people building with ai agents have never built one here's what it actually is: → instructions the agent reads before touching anything → state that persists so it never starts from zero → verification gates it can't skip to declare done → scope that locks it to one feature at a time → a session lifecycle so every run starts clean and ends clean without this, your agent writes code, says "done," and breaks everything. with this, it picks up where it left off, finishes what it started, and proves it before moving on learn-harness-engineering is a free course built around exactly this 12 lectures. 6 hands-on projects. one real app that evolves as your harness skills grow if you're using claude code or codex on real work and the output still feels unreliable now you know why github.com/walkinglabs/le…

[🚨 POR FAVOR AYÚDAME A DIFUNDIR — BUSCO UN ENSAYO CLÍNICO EN CUALQUIER PARTE DEL MUNDO 🌍] Mi nombre es Selva Alvarez, tengo 35 años, vivo en Colombia y soy Científica de Datos 📊. Hoy no les escribo solo como una paciente con cáncer gástrico avanzado (Estadio IV) 🎗️, sino como alguien que acaba de encontrar un dato científico único en su propio cuerpo que podría cambiarlo todo 🧬. Ayer, 21 de mayo de 2026, mi reporte de patología reveló algo extraordinario y muy escaso en el mundo: mi tumor dio POSITIVO para una proteína llamada CLAUDINA 18.2 🔬. Para explicarlo de forma sencilla: la Claudina 18.2 es como una "cerradura" 🔐 muy rara que solo tienen algunos tumores en el mundo. La ciencia médica ya inventó una "llave" 🔑 exacta para esa cerradura (un medicamento avanzado llamado Zolbetuximab o Vyloy), que es capaz de atacar directamente las células enfermas sin destruir el resto del cuerpo 🎯. El problema es que este medicamento no es comercial en Colombia 🇨🇴. Mi oncólogo me confirmó que mi única oportunidad real de acceder a esta "llave" es que una farmacéutica o un centro de investigación me reciba en un Ensayo Clínico Nacional o Internacional o en un programa de acceso expandido ✈️. Al ser un perfil de paciente tan escaso y específico en el planeta, sé que hay laboratorios y científicos en el mundo buscando activamente a personas como yo para sus estudios de medicina de precisión 🩺. Quiero que el mundo sepa que estoy lista 💪: mis órganos vitales (hígado y riñones) están completamente sanos, fuertes y funcionando a la perfección, listos para resistir el tratamiento 🫁. Tengo la juventud, la fuerza biológica y la determinación absoluta de pelear por mi vida y aportar a la ciencia 🦾. Hago un llamado directo a la comunidad científica global (#OncoTwitter 🩺), a laboratorios como @AstellasUS 🏢, y a investigadores de cualquier país que lideren estudios sobre Claudina 18.2: Aquí hay una paciente joven con el biomarcador idóneo 📑. Por favor, ayúdenme con un RT (compartir) 🔁. Un solo clic de ustedes puede hacer que este mensaje cruce fronteras y llegue al comité científico o a la farmacéutica que me pueda patrocinar 💌. Mis mensajes directos (DM) están abiertos para enviar de inmediato mi historial médico, resultados de los marcadores, bloques de patología o cualquier cosa que se necesite 📤. Colombia 🇨🇴 | Contacto: DM abierto 📩. #GastricCancer #Zolbetuximab #ClinicalTrials #PrecisionMedicine #OncoTwitter #Claudina18 #CancerGastrico #EnsayosClinicos #MedicinaDePrecision #Vyloy

Andrej Karpathy just explained the future of software engineering without directly saying it. The best AI engineers are no longer “prompting.” They’re building systems around the agents. Karpathy’s biggest insight wasn’t: “Claude can code.” It was: LLMs become dramatically better when you force them into disciplined workflows. That’s why "CLAUDE.md" files are suddenly everywhere. Not because they’re prompts. Because they behave like an operating system for the agent. Karpathy called out the exact problems with AI coding: - models assume instead of asking - they overengineer simple tasks - they hide confusion - they rewrite unrelated code - they optimize for completion, not correctness So developers started encoding rules directly into the workflow: → Think before coding → Simplicity first → Surgical edits only → Goal-driven execution And the results are wild. People are now running multiple Claude Code agents in parallel like engineering teams: • one agent researching • one debugging • one writing tests • one optimizing code • one validating outputs Not “AI assistance.” Actual orchestration. And this part from Karpathy changes everything: “Don’t tell the model what to do. Give it success criteria and let it loop.” That is the shift. From: “write this function” To: “here’s the goal, constraints, tests, and verification system — now iterate until correct.” The craziest part? This already feels like a phase shift in engineering. A lot of developers quietly went from: 80% manual coding → to 80% agent-driven coding in just months. Not because AI became perfect. Because the leverage became impossible to ignore. We’re entering an era where the highest leverage engineers won’t necessarily be the best coders. They’ll be the people who build the best systems around AI agents.

🚨 ANTHROPIC JUST KILLED THE DEMO AGENT ERA. Their Agents team showed exactly what production grade looks like. Not theory. Not a tutorial. A four layer framework for multi agent systems built to actually work in the real world. 30 minutes. This is the video I wish existed 6 months ago.

Microsoft Senior AI developer just showed how they build AI agents with Claude at Microsoft. 34-minutes. free. By Microsoft team Opus 4.7 + 1,400+ pre-built MCP tools plug Claude into agent → give it tools → ship to production worth more than any $500 vibe-coding course.

@nicos_ai @G_Programming Alan. Este problema es el mismo que se soluciona con Gentle AI usando los comandos de SDD? 🙏🏽🙏🏽

GitHub acaba de solucionar el mayor problema del vibe coding. Acaban de lanzar Spec Kit y en días ya tiene +95K estrellas. ¿La idea? En vez de tirar prompts vagos y rezar para que el agente no rompa tu proyecto… Spec Kit obliga a la IA a crear una especificación estructurada ANTES de tocar código. La IA primero entiende lo que quieres construir, pregunta lo que falta, organiza el proyecto y después empieza a programar. Eso significa menos tiempo arreglando errores absurdos, menos código inconsistente y resultados mucho más predecibles cuando trabajas con agentes. El flujo es simple: /constitution → reglas y estándares /specify → qué quieres construir /clarify → dudas antes de empezar /plan → arquitectura y stack /tasks → tareas ordenadas /implement → ejecución Compatible con Claude Code, Cursor, Copilot, Codex, Gemini CLI y +25 agentes. 95K estrellas. 8K forks. Open source. Publicado por GitHub. Repositorio 👇

Andrej Karpathy: "90% of Claude's mistakes come from missing context, not a weak model." 41% mistake rate without a CLAUDE.md. 11% with the 4-rule baseline. 3% with the 12-rule version below here are the 12 rules senior engineers settled on: 1. think before coding: state assumptions, don't guess. the model can't read your mind, stop hoping it will 2. simplicity first: minimum code, no speculative abstractions. the moment you let Claude add "for future flexibility," you've added 200 lines you'll delete next quarter 3. surgical changes: touch only what you must. don't let it improve adjacent code, that's how PRs blow up 4. goal-driven execution: define success criteria upfront, loop until verified. without them Claude either loops forever or stops too early 5. use the model only for judgment calls: classification, drafting, summarization, extraction. NOT routing, retries, status-code handling, deterministic transforms. if code can answer, code answers 6. token budgets are not advisory: per-task 4000, per-session 30000. by message 40 of a long debug, Claude is re-suggesting fixes you rejected at message 5 7. surface conflicts, don't average them: two patterns in the codebase? pick one. Claude blending them is how errors get swallowed twice 8. read before you write: read exports, callers, shared utilities. Claude will happily add a duplicate function next to an identical one it never read 9. tests verify intent, not just behavior: a test that can't fail when business logic changes is wrong. all 12 of Claude's tests can pass while the function returns a constant 10. checkpoint every significant step: Claude finished steps 5 and 6 on top of a broken state from step 4. nobody noticed for an hour 11. match the codebase conventions: class components? don't fork to hooks silently. testing patterns assumed componentDidMount, hooks broke them without surfacing 12. fail loud: "completed successfully" with 14% of records silently skipped is the worst class of bug. surface uncertainty, don't hide it what actually compounds instead of the next framework: - the CLAUDE.md file as institutional memory across sessions - eval-driven changes, not vibe-driven - checkpoints over speed - explicit conflicts over silent blending - discipline over framework, every time - one repo, one rules file, no exceptions be a few rules ahead of AI twitter before this becomes mass-opinion study this

@aidenybai You need to solve UI architecture before you can solve UI testing. Foldkit Scene tests are pure, fast, and test what you actually care about. No browser, no JSDOM, no mocks. foldkit.dev/testing/scene

A guy ran Karpathy's CLAUDE md across 30 codebases for 6 weeks. 🚨 Claude's mistake rate went from 41% → 11% Then he added 8 more rules and got it to 3% The rules are actually fire (even if the numbers are sus)

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

Claude Code feels completely different once you install this. Anthropic quietly released an official plugin called claude-code-setup and it basically turns Claude Code from “pretty good” into an actual AI dev environment. It scans your project and recommends: → hooks → skills → MCP servers → subagents → automations Then sets everything up step-by-step for you. Most people are using Claude Code completely vanilla… which is why their experience feels messy. The real power comes from the ecosystem around it. Install: /plugin install claude-code-setup@claude-plugins-official Bookmark this before you forget it.

Harness Engineering A Design Guide to Claude Code drive.google.com/file/d/1mIuy2k…

This is the best site on the internet to learn harness engineering walkinglabs.github.io/learn-harness-…

Is anyone doing feature flag development with agents? Not tried it, but in theory feature flagging is an alternative model to PR's to getting work on main. 1. Put it on main, disabled by a flag 2. Deploy with the rest of the system 3. Unflag to selected users early 4. Fix bugs for those users 5. Unflag to more users 6. Repeat until shipped Feels like a perfect strategy to pair with agents

RT @trq212: a prompt I've been using a lot recently: implement <SPEC> and while you do, keep a running implementation-notes.html file (or…

Este artículo de @addyosmani me encantó: "no delegues tu aprendizaje a la AI" Acerca de varios estudios, incluso algunos hechos por Anthropic, dónde analizan que los devs que usan la AI solo generar código sin analizarlo ni razonarlo. En estos estudios, solo el 40% de los