How to Evaluate Agent Retrieval
Glean details their rigorous process for evaluating AI retrieval systems, covering custom eval set creation from historical usage and synthetic data, and advanced metrics like Recall@K and agentic judges.
Read articleTopic
Independent digests of AI Evals Cohort talks — key takeaways, technical details, and related reading.
Suggested route
Move from earlier context through the available material to the newest signals. This route is a reading aid, not a difficulty ranking.
Glean details their rigorous process for evaluating AI retrieval systems, covering custom eval set creation from historical usage and synthetic data, and advanced metrics like Recall@K and agentic judges.
Read article
Learn the core concepts of AI Evals, differentiating between general model benchmarks and product-specific testing. We cover tracing, error analysis, and setting up automated, robust evaluation pipelines.
Read article