AI Engineer · October 17, 2026

Agent calculations: trace assumptions, code changes, and review decisions

video AI Engineering Model Evaluation Agent Security

Agent calculations: trace assumptions, code changes, and review decisions video thumbnail

Why it matters

Francesco Galletta and Felippe Felisola Caso use a wedding-budget demo to connect tested calculations with evidence lineage. Inputs are distinguished as sourced facts, assumptions, or interpretations, and reviewer challenges feed back into the plan. Repeatable arithmetic can still use wrong assumptions or changed code. The speakers present a design example in a personal capacity, not a verified Monzo deployment.

My takeaway: Use synthetic data to change an assumption and the calculation code independently. Verify each result records its inputs, source, code version, and approval, and that dependent results are recalculated. Compare with an independently checked example. Record whether a reviewer challenge changed the plan or was dismissed, with a reason.

Keep exploring

More curated notes connected through AI Engineering and Model Evaluation.

OpenAI News · framework

Pacing model development in an era of cyber-critical capabilities

OpenAI says preliminary evidence that Astra may meet its Critical cybersecurity threshold led it to pause frontier reinforcement-learning work for two weeks and keep its largest planned run on hold. New safeguards include stronger workload and network isolation, continuous boundary testing, token-level monitoring that escalates suspicious tool activity, and broader alignment checks for deception, reward hacking, and unauthorized access.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.

OpenAI News · framework

A blueprint for democratic governance of frontier AI

OpenAI proposes a three-part U.S. frontier-AI governance model: harmonize emerging state safety laws into a federal baseline, strengthen CAISI as an evaluation and standards institution, and coordinate a broader resilience program. Proposed controls include severe-risk evaluations, transparency reports, independent audits, safety-incident reporting, model-weight security, whistleblower protection, and periodic technical assessments.