Keynote

Beyond the API: Modern Inference for Modern Workloads

Why fine-tuning is quietly resurging as "RL," how teams are compressing inference costs with smaller specialized models, and who owns the model routing problem.

Speakers