Essays on real work, AI agents, supervision, and the conditions for trust in work systems.https://julientalbot.comen-ushttps://julientalbot.com/en/writing/ia-deplace-le-travail-reelhttps://julientalbot.com/en/writing/ia-deplace-le-travail-reelWhy an AI project does not merely remove tasks: it moves workload, exceptions, judgment, responsibility, and sometimes the meaning of work.Mon, 22 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/ai-agents-are-operatorshttps://julientalbot.com/en/writing/ai-agents-are-operatorsBenchmarks test models. Real work tests situated operators: model, harness, tools, permissions, evidence, recovery, and human belief.Fri, 19 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/making-grok-acthttps://julientalbot.com/en/writing/making-grok-actAn XML system block that forbids intent narration: one repro (10 min → 33.9 s), corpus measures, an honest A/B, and what the patch cannot fix.Fri, 12 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/enterprise-agent-problem-is-beliefhttps://julientalbot.com/en/writing/enterprise-agent-problem-is-beliefAn agent is not enterprise-ready because it can act. It is enterprise-ready when the belief it creates about its action is calibrated to evidence.Thu, 04 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/agents-ia-pme-reunion-travail-reelhttps://julientalbot.com/en/writing/agents-ia-pme-reunion-travail-reelA useful AI agent does not start with a technical demo. It starts with a situated task, limited rights, verifiable evidence, and a possible human recovery path.Tue, 02 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/conference-ia-travail-reelhttps://julientalbot.com/en/writing/conference-ia-travail-reelWhy a useful AI talk should start from what AI changes in work, decisions, responsibility, evidence and human recovery, not from tools alone.Tue, 02 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/duerp-tms-rps-qvct-travail-reelhttps://julientalbot.com/en/writing/duerp-tms-rps-qvct-travail-reelA DUERP becomes useful in Réunion when it connects risks to observable situations, production constraints, and the real room people have to maneuver.Tue, 02 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/ergonomie-architecte-appel-offres-reunionhttps://julientalbot.com/en/writing/ergonomie-architecte-appel-offres-reunionWhat an architect, owner-assistance team, or project manager can ask from an ergonomist to strengthen a proposal, a program, or a reconstruction.Tue, 02 Jun 2026 00:00:00 GMThttps://julientalbot.com/en/writing/raw-traces-are-not-evalshttps://julientalbot.com/en/writing/raw-traces-are-not-evalsThe missing layer between real agent failures and measurable model progress: reducing messy traces into replayable eval seeds without laundering the signal.Fri, 29 May 2026 00:00:00 GMThttps://julientalbot.com/en/writing/tool-use-is-not-task-completionhttps://julientalbot.com/en/writing/tool-use-is-not-task-completionWhy agent reliability depends on preserving the boundary between intention, action, observable state, and truthful final claims.Tue, 19 May 2026 00:00:00 GMThttps://julientalbot.com/en/writing/claim-action-evidencehttps://julientalbot.com/en/writing/claim-action-evidenceWhy the critical risk is not only hallucination, but an agent claiming work is done without observable evidence.Thu, 14 May 2026 00:00:00 GMThttps://julientalbot.com/en/writing/agents-are-work-systemshttps://julientalbot.com/en/writing/agents-are-work-systemsWhy task completion is not enough: an agent redistributes verification, responsibility, coordination, and recovery work.Sun, 03 May 2026 00:00:00 GMThttps://julientalbot.com/en/writing/ai-reconfigures-workhttps://julientalbot.com/en/writing/ai-reconfigures-workWhy AI must be analyzed from real activity: constraints, trade-offs, invisible cooperation, room to act, and responsibility.Fri, 24 Apr 2026 00:00:00 GMThttps://julientalbot.com/en/writing/benchmark-liehttps://julientalbot.com/en/writing/benchmark-lieA cognitive ergonomist dogfoods Grok 4.20 across 12+ production agents. What 40 years of human factors research says about what benchmarks miss in agent loops.Tue, 21 Apr 2026 00:00:00 GMThttps://julientalbot.com/en/writing/ergonomie-ia-ouverturehttps://julientalbot.com/en/writing/ergonomie-ia-ouvertureOpening note. What AI concretely displaces in the ergonomist's practice and in organizational work — and the analysis space it opens.Sat, 18 Apr 2026 00:00:00 GMT