Context Engineering: How to Build Valuable AI Products
What’s the difference between an AI product that feels magical and one that feels clunky? It’s context engineering, aka deciding which 20,000 tokens your LLM actually needs. Search, retrieval, memory, personalization, and tool calls all determine how relevant and useful your AI’s answers and actions will be.
That’s why we’re bringing together three companies tackling this from different angles:
Ghita from ZeroEntropy – builds the most accurate retrieval models, specializing in rerankers to optimize your stack.
Prateek from Mem0 – provides the memory layer for AI agents, making them more personalized and useful over time.
Abhi from Mastra– a TypeScript framework for AI agents, enabling smarter tool calls and seamless source–LLM connections.
Each will give a 10-minute talk diving into practical strategies for context engineering: from when to use rerankers, to how to design memory for agents, to connecting the right tools to the right models.
Who should attend? Developers and AI leaders building products who care about accuracy, speed, cost-efficiency, and great user experience.
If you want to understand how to make your AI product not just work, but truly deliver value, this event is for you.
-> Stay in touch through the official Context Engineers community here.