
X Article
Excerpt — full reading on X
Hermes HUD : quand l’agent vient se poser sur l’écran
Les assistants IA savent expliquer. Ils savent moins souvent montrer.
Members’ publications
Articles published on X by Lab members. Read an excerpt here, then open the source for the full text.
Signals collect shorter posts. This page brings together long-form articles; group discussions and the wiki remain in the members’ space.
Catalogue dated 18 November 2026 (UTC). Some excerpts come from earlier collections.
26 articles · page 2 of 3

X Article
Excerpt — full reading on X
Les assistants IA savent expliquer. Ils savent moins souvent montrer.

X Article
Excerpt — full reading on X
Codex prepares the context, Oracle sends a controlled bundle to GPT‑5.6 Sol in ChatGPT Pro, and Codex reviews it before coding and testing. Roles stay separate: Pro reasoning on one side, Codex execution on the other—without the OpenAI API or sending the repository blind.
Translated from French.

X Article
Excerpt — full reading on X
Grok 4.6 offers a clear price-performance trade-off. At the same CursorBench score, it costs six times less than Fable 5 Max. It suits well-scoped execution better than implicit judgment calls. I delegate execution; I keep the decision-making.
Translated from French.

X Article
Excerpt — full reading on X
I repurposed a 2018 Mac mini as a Linux server for an always-on AI agent. Between the T2 chip and Ubuntu/macOS dual boot, this home-lab project raised challenges around private access, tech monitoring and automation. Here is the full technical field report.
Translated from French.

X Article
Excerpt — full reading on X
Opus 5 needs fewer redundant checks in the prompt and more evidence captured by the system. The model already self-checks—demand real execution traces, not claims.
Translated from French.

X Article
Excerpt — full reading on X
Kimi K3 does not beat Claude Fable 5 or GPT‑5.6 Sol, but cuts the cost per task to roughly one-third of Fable 5's. With 2.8 trillion parameters and open weights coming soon, it prioritizes efficiency over the top spot.
Translated from French.

X Article
Excerpt — full reading on X
Field report: no LLM is universally best. Claude Fable 5 is impressive for development but remains expensive, while GPT‑5.6 Sol is my default because its persistence on long tasks offsets the risk of drift.
Translated from French.

X Article
Excerpt — full reading on X
Planka + Hermes Kanban turns a board into a shared workspace for humans and AI agents. Cards become work units that anyone can read, edit and execute, with better coordination, durable context and clear handoffs.
Translated from French.

X Article
Excerpt — full reading on X
L'IA ne rend pas ton Obsidian plus intelligent. Elle révèle à quel point il est flou.

X Article
Excerpt — full reading on X
At Build 2026, Microsoft's ambition goes beyond software: a complete operating system for the agentic era. Its integrated stack spans models, agents, cloud, hardware and energy. Alongside NVIDIA, Microsoft aims to become an industrial AI operator as compute demand explodes.
Translated from French.

X Article
Excerpt — full reading on X
AMI Labs raises $1.03 billion to industrialize world models. The goal: AI that understands the world and predicts the consequences of its actions, beyond language alone. Yann LeCun's bet on robotics and industry.
Translated from French.

X Article
Excerpt — full reading on X
AI is not immaterial: every token consumes compute and energy. Mistral is competing not only on models, but across the value chain. Europe must connect industry, energy and regulation to build digital sovereignty.
Translated from French.