The Fork Bomb Saved Us
The LiteLLM attacker made one implementation mistake. That mistake was the only reason 47,000 installs got caught in 46 minutes.
AI & TechnologyEcriture
Reflexions sur le design, le developpement, l'infrastructure IA et la creation de produits.
The Agent Didn't Get Smarter
The model is the same between session 1 and session 500. The project changed. This reframes the entire AI productivity conversation.
AI & TechnologyOvernight
Between midnight and 6am, Googlebot crawls 21,000 pages, Bingbot crawls 10,000, and the comprehensive check grinds through 15,000. The site is more active at 3am than at 3pm.
AI & TechnologyThe Evidence Gate
I believe and it should are not evidence. Every completion report needs a file path, test output, or specific code. The discipline of proof in an age of plausible AI output.
AI & TechnologySeventeen Thousand Signals
My vault has 17,900 signals from arXiv, Semantic Scholar, HN, NVD, and 9 other sources. Most are noise. The noise taught me what signal looks like.
AI & TechnologyWhat I Run Before I Sleep
Every night: 15,000 pages checked, TTFB measured, cache verified, sitemaps crawled. The goodnight routine is where operational discipline lives.
AI & TechnologyThe Handoff Document
A diagnosis that survived three code review corrections, two priority reorderings, and guided the correct implementation four days later. The most underrated artifact in agent-assisted development.
AI & TechnologyQuality Is the Only Variable
Time, cost, resources, and effort are not constraints. The question is what's right, not what's efficient. A philosophy for building with AI agents.
AI & TechnologyTaste Is Infrastructure
As agents generate more of what ships, the quality ceiling is set by how well you encode aesthetic judgment into systems. Taste scales when it becomes queryable.
AI & TechnologyWhen Your Agent Becomes the Researcher
Claude Code discovered adversarial attacks that outperform 30+ existing methods. Autoresearch just crossed from experiment to capability.
AI & TechnologyCompound Context: Why AI Projects Get Better the Longer You Stay With Them
Every problem you solve with an AI agent deposits context that the next session withdraws with interest. This is context compounding.
AI & TechnologyThe Supply Chain Is the Attack Surface
Trivy got compromised. Then LiteLLM. Then 47,000 installs in 46 minutes. The AI supply chain worked exactly as designed.
AI & TechnologyPublications externes
Redaction technique chez Introl
Local LLM Hardware Pricing Guide 2025
Recommandations matérielles complètes et analyse des coûts pour l'exécution locale de grands modèles de langage.
H100 vs. H200 vs. B200: Choosing the Right NVIDIA GPUs for Your AI Workload
Guide de sélection GPU comparant les derniers accélérateurs de datacenter NVIDIA pour différentes charges de travail IA.
TPU Architecture: Complete Guide to Google's 7 Generations
Analyse technique approfondie de l'évolution du Tensor Processing Unit de Google, de TPUv1 à TPUv5.
GPU Memory Pooling in Multi-Tenant Kubernetes
Stratégies de partage de ressources pour les clusters GPU dans les environnements conteneurisés.
Ray Clusters: Distributed AI Computing Infrastructure
Guide pour construire et gérer le calcul IA distribué avec le framework Ray.
DeepSeek V3 Open Source AI Cost Advantage
Analyse de l'économie des LLM open source et du positionnement concurrentiel de DeepSeek.
NVIDIA Vera Rubin GPU 600kW Racks 2027
Besoins énergétiques futurs des datacenters et feuille de route GPU de nouvelle génération de NVIDIA.
SMR Nuclear Power for AI Data Centers
Solutions de petits réacteurs modulaires pour alimenter l'infrastructure IA de nouvelle génération.
DeepSeek MHC Architecture Breakthrough
Analyse technique des innovations d'architecture Multi-Head Compression de DeepSeek.