Session catalog
Filter and search across every session at Atlas Conference 2026.
1 filter applied
5 sessions match your filters.
Tuesday, June 9, 2026
2 sessionsAI Safety & Agent EvaluationAlmost Full
LLM-as-a-Judge for Enterprise AI Evaluation
When and how to trust LLM judges. We discuss calibration, rubric design, jury-of-models patterns, and pitfalls when using models to evaluate other models in production.
- When
- Tue, Jun 9 · 10:00 AM – 11:00 AM
- Where
- Hall D1
- Speaker
- Dr. Yuki Tanaka · Polaris Research
- Level
- Advanced
230 / 240 registered10 seats left
AI Safety & Agent EvaluationAvailable
Building Evaluation Datasets that Reflect Reality
From production traces to high-signal eval sets: sampling, labeling pipelines, and avoiding eval drift over time.
- When
- Tue, Jun 9 · 4:30 PM – 5:30 PM
- Where
- Hall D1
- Speaker
- Ingrid Sørensen · Polaris Research
- Level
- Intermediate
89 / 180 registered91 seats left
Wednesday, June 10, 2026
2 sessionsAI Safety & Agent EvaluationAlmost Full
Red Teaming Agentic Systems
Adversarial evaluation strategies for agents with tool access: prompt injection, goal hijacking, exfiltration tests, and automated attack libraries.
- When
- Wed, Jun 10 · 10:00 AM – 11:00 AM
- Where
- Hall D2
- Speaker
- Nadia Hassan · Bastion Security
- Level
- Advanced
199 / 200 registered1 seats left
AI Safety & Agent EvaluationAvailable
Measuring Agent Reliability in Long-Running Tasks
How to design metrics that capture multi-step success, partial credit, and recovery behavior for agents that run for hours.
- When
- Wed, Jun 10 · 3:30 PM – 4:30 PM
- Where
- Hall D1
- Speaker
- Ben Carter · Sentinel AI
- Level
- Intermediate
88 / 160 registered72 seats left