Achieving Accuracy and Efficiency in AI Agents

​An evening of food, drinks, and technical discussion focused on how to achieve both efficiency and accuracy in AI agents.

​
The talks will cover:

• ​Stop “Vibe Testing” Your LLMs. It’s Time for Real Evals by Sara Wiltberger, Senior PM @Google Labs

• ​GEPA: Reflective Prompt Optimization by Lakshya A. Agrawal, Creator of GEPA @Sky Lab, UC Berkeley

• ​Hardware Trends for LLM Inference by Oto (Inference Lead @Dria)

• ​Efficient LLM Serving by Deep Dhillon (SWE @Modular)

​Agenda

• ​17:45–18:15: Welcome & pizza

• ​18:15–19:15: Lightning sessions

• ​19:15–20:00: Networking

​Whether you're experimenting with LLMs, building eval frameworks, or designing agentic workflows, this event brings together the researchers, engineers, and founders who are shaping the future of efficient, reliable AI.