AI Engineer · October 19, 2026

Model swaps: test tool arguments and calibrate the evaluator

video Model Evaluation AI Engineering

Model swaps: test tool arguments and calibrate the evaluator video thumbnail

Why it matters

Pamela Fox’s workshop compares models on application tasks and turns calendar-tool behavior into checks of the selected tool and returned arguments. For less exact outputs, it records context and uses a judge that must itself be calibrated. Generated scenarios can bootstrap coverage, but a small demo score cannot establish which model is best for a production workload.

My takeaway: Create cases for date boundaries, ambiguous durations, supported languages, and missing tool steps. Assert exact tool arguments where possible. Save inputs, retrieved context, and tool traces so you can change the judge without rerunning the app. Compare judgments with expert labels, add real failures, and keep prompt-tuning examples separate from held-out model comparisons.

Keep exploring

More curated notes connected through Model Evaluation and AI Engineering.

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.

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.

OpenAI News · framework

OpenAI’s Frontier Governance Framework

OpenAI's 22-page Frontier Governance Framework maps its frontier-model processes to California's Transparency in Frontier AI Act and the EU AI Act's general-purpose AI code. It documents lifecycle risk assessment, cyber-offense and other risk tiers, mitigation and residual-risk decisions, critical-incident handling, security risk management, model reporting, external review, responsibility allocation, and change control.