Aditya Palnitkar
Staff Software Engineer at Meta
Aditya is a staff engineer / TL at Meta. After his undergrad from BITS Pilani, he graduated with a MS in CS from Stanford, with a focus on using ML for performing graph analysis and predictions. In his decade of working in AI / ML, he has gained unique experience and insights working on state of the art recommendations models backing Facebook Reels and Facebook Watch, which serve 1B+ users every day. He is now tech-leading a team working on evaluating AI agents being built at Meta.
All Sessions by Aditya Palnitkar
Virtual Only: Evals for Supercharging your AI Agents
<span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="secfocus">Beginner-Intermediate</span> </span>Deploying LLM applications without robust testing and monitoring is a recipe for failure. In this talk, you'll discover how a world-class evaluation system can supercharge your AI agents by turning potential fragility into development velocity. Learn to build rigorous feedback loops that drive ROI—from creating high-quality, human-labeled datasets and leveraging LLMs as both judges and user simulators, to designing benchmarks that catch hallucinations and monitor failure modes. Join us to unlock strategies that align evaluation metrics with your business goals, ensuring your LLM applications not only perform reliably but continuously improve.
Virtual Only: Evals for Supercharging your AI Agents
<span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="secfocus">Beginner-Intermediate</span> </span>Deploying LLM applications without robust testing and monitoring is a recipe for failure. In this talk, you'll discover how a world-class evaluation system can supercharge your AI agents by turning potential fragility into development velocity. Learn to build rigorous feedback loops that drive ROI—from creating high-quality, human-labeled datasets and leveraging LLMs as both judges and user simulators, to designing benchmarks that catch hallucinations and monitor failure modes. Join us to unlock strategies that align evaluation metrics with your business goals, ensuring your LLM applications not only perform reliably but continuously improve.
