Nina Lopatina, PhD, is a Staff Developer Advocate at MongoDB, where she helps developers build AI applications and agentic systems with the right data, context, and architecture. She focuses on turning complex topics like agent workflows, search, and retrieval-augmented generation (RAG) into clear technical content, hands-on demos, and education that help developers build with confidence. Much of her current work centers on memory and context engineering for AI agents, the throughline of her Ai4 session, "Memory Systems for AI Agents," which gives developers a practical framework for deciding what an agent should remember, where to persist it, and what to retrieve at runtime so agentic systems hold up in production. Before joining MongoDB, Nina led Developer Advocacy at Contextual AI, where she helped developers build accurate, scalable RAG agents and agentic search solutions. Her background spans machine learning, NLP, and language modeling, and she began her career applying ML techniques to neural data during her PhD and postdoctoral research in reinforcement learning and decision-making. That research foundation shapes how she approaches developer education today, grounding practical guidance in how these systems actually behave. Off the clock, you'll find Nina lapping moguls or climbing mountains, depending on the season.