Skip to content

Reuse knowledge for Agentic AI: Workshop on Agent Memory, lowering LLM costs, and reducing latency

with Bhavana Giri & Ashwin Hariharan

Thursday 26 November 15:10 – 17:10 Room 9

About This Session

Imagine your pager going off at 3 a.m. because your LLM costs are about to exceed their limit, all while your AI agent keeps forgetting important context. Generative AI applications can quickly become slow, expensive, and frustrating when every request is sent directly to an LLM. In this hands-on workshop, you’ll build a full-stack AI application that is faster, more cost-efficient, and capable of remembering what matters. You’ll learn how to: - Maintain conversational continuity and context. - Identify important details from conversations and preserve them in long-term memory. - Implement semantic caching so valuable answers can be reused for similar questions.

Topics

  • Agentic AI
  • Best Practices
  • Generative AI (GenAI)
  • Redis