Thursday, 13 August 2026 at 17:00
Quality Court, London, WC2A 1HR, Quality Court
## Details By now it’s standard advice, in the research and in practice: don’t let a model grade its own homework, bring in a second one to review it. It does find things. But it reviews inside the frame you handed it, and if it can detect what answer you’re hoping for, it tends to find its way there. This is a walkthrough of a real Claude Code conversation where both of those bit me on my own work, and what I did about each. By that point Carmen Mardiros had a specific understanding of how sycophancy, autoregressive commitment, and recency bias behave, and used it to steer the conversation while it was running. That steering is most of what I want to show, so that other people can leverage the same mechanisms. It may change how you run reviews with the current generation of frontier models. What you’ll leave with: 🎯 What to withhold from a reviewing agent so it challenges the premise rather than the detail 🎯 What to say to an agent that’s already agreed with you without triggering over-correction in the other direction 🎯 How to ask for evaluation, opinion and conclusions without leaking your implied bias to the model. 🗓️ Save the Date: 13th August, 2026 📍 Location: Sahaj Software, 1 Quality Court, London, WC2A 1HR 🕕 Time: 6:00 PM – 8:00 PM (BST) The evening will feature: 🎤 One engaging talk with Q&A 🤝 Plenty of time for networking over free pizza and non-alcoholic drinks Don’t miss this chance to learn, connect, and share with fellow tech enthusiasts. 👉 Please click the link below to register. https://sahaj.ai/events/how-to-make-agentic-adversarial-review-actually-adversarial/
View event details →Listed from meetup. Details can change — always check with the organiser.