This event has ended. Sign up now to watch on-demand.
Build and monitor LLM applications with confidence. As teams adopt large language models, speed and reliability are critical–but so are cost, latency, and user experience.
Join us to see how Datadog LLM Observability helps AI engineers, data scientists, and developers:
- Test prompts and models rapidly in the Playground: Adjust inputs, swap models, and fine-tune parameters in minutes
- Run structured experiments: Compare prompts and models against production datasets, catch regressions, and balance performance, quality, and cost
- Trace chains and agentic workflows end to end: Detect latency, error, and token usage issues across every step
- Evaluate accuracy and safety: Apply out-of-the-box and custom checks for hallucinations, prompt injections, and PII exposure
- Monitor live performance at scale: Correlate LLM metrics with application, infrastructure, and user experience data across Datadog
Charles Jacquet
Product Manager, Datadog
Will Potts
Senior Product Marketing Manager, Datadog