Recent posts

Don't miss the latest trends

GitHub Copilot Fusion: You Stop Picking Models, and That Changes More Than Your Bill

Short version for busy readers. On October 28, 2026, at GitHub Universe, GitHub made Copilot Fusion generally available. You no longer pick Claude, GPT or Gemini in the model picker. You pick “Copilot Fusion” plus an effort level (Standard, High or Max), and Copilot decides, prompt by prompt, which models run and how they cooperate:…

SAMI
November 7, 2026
Artificial Intelligence

Top LLM Cost Optimization Techniques Every Engineer Should Know

Building an AI feature has never been easier. A working prototype takes an afternoon: pick a frontier model, write a prompt, wire up a few tools, ship it. Then the invoice arrives. The prototype that cost a few dollars a day in testing now serves thousands of users. Every request carries a 6,000-token system prompt.…

22 mins to read
Artificial Intelligence

Chatbot ou agent IA : où se situe vraiment votre système sur l’échelle d’autonomie

Chatbot ou agent IA Si vous avez assisté cette année à une démo d’un éditeur IA dans une banque, vous avez sans doute vu la slide : « Découvrez votre premier agent IA ». Puis la démo commence. L’« agent » répond à trois questions sur les frais de tenue de compte à partir d’un…

17 mins to read
Artificial Intelligence

Chatbot vs AI Agent: Where Your System Really Sits on the Autonomy Spectrum

Chatbot vs AI Agent If you have sat through an AI vendor demo in a bank this year, you have probably seen the slide. “Meet your first AI agent.” Then the demo starts, and the agent answers three questions about account fees from a product PDF and offers a link to reset a password. It’s…

15 mins to read
Artificial Intelligence

Arrêtez de récupérer, commencez à compiler : le pattern LLM Wiki

Le RAG est un interpréteur, le LLM Wiki est un compilateur Le RAG nous a habitués à une drôle de pratique : chaque question repaie intégralement le prix de la compréhension du corpus. Le pattern LLM Wiki refuse de payer ce prix au moment de la requête. Il déplace la synthèse vers l’ingestion : un…

42 mins to read
Artificial Intelligence

Stop retrieving, start compiling: the LLM Wiki pattern

RAG is an interpreter. The LLM Wiki is a compiler RAG trained us to accept a strange habit: every question pays the full price of understanding the corpus again. The LLM Wiki pattern refuses to pay that price at query time. It moves synthesis to ingestion, where an agent reads each source once, writes what…

34 mins to read
Artificial Intelligence

From factory to intelligence : The evidence plane, and the asset that compounds

Post 4 of 8 in the series on rebuilding a bank’s software development factory for the agentic era. Here is the architecture. Five layers, two cross-cutting planes, and one workstream that decides whether any of the rest of it works. I will start at the bottom, because the bottom is the part that gets cut…

10 mins to read
Artificial Intelligence

From factory to intelligence : What survives contact with a regulator

Post 3 of 8 in the series on rebuilding a bank’s software development factory for the agentic era. Block’s “From Hierarchy to Intelligence” is the most interesting organisational design published this year, and roughly half of it is illegal to implement in a supervised European credit institution. That sounds like a complaint. It is not.…

9 mins to read
Artificial Intelligence

From factory to intelligence : The AI token bill

Post 2 of 8 in the series on rebuilding a bank’s software development factory for the agentic era. The AI token bill : This is the post I would have wanted to read eighteen months ago, so I am putting it second rather than burying it in the middle of a design series. Here is…

11 mins to read
Artificial Intelligence Software Design and Architecture

L’usine de développement logiciel fonctionne toujours, c’est la contrainte qui a bougé

Article 1 sur 8 de la série consacrée à la refonte de l’usine de développement logiciel d’une banque à l’ère des agents. Il y a 3 ans, j’ai publié ici un long guide sur la conception d’une usine de développement logiciel pour une banque. 9 chantiers, Team Topologies, les capacités Accelerate, les métriques DORA, une…

12 mins to read