Reducing Agent Token Costs + RAG Beyond Semantic Search
Why agentic systems burn tokens, what retrieval misses, and how structured knowledge changes the runtime economics.
Interview with Roie Schwaber-Cohen.
- Agents
- Token economics
- RAG
Conference talks, podcasts, panels, and working sessions about search, knowledge, agent systems, and the production reality behind them.
Official host pages, recordings, and transcripts.
Why agentic systems burn tokens, what retrieval misses, and how structured knowledge changes the runtime economics.
Interview with Roie Schwaber-Cohen.
A hands-on session on query routing, applicability, Pinecone Assistant, and n8n for customer questions that span multiple domains.
Code-along presented by Roie Schwaber-Cohen and Jenna Pederson.
A long-form conversation about contextual truth, applicability, messy enterprise knowledge, and the infrastructure agents actually need.
Interview with Roie Schwaber-Cohen.
A discussion with Amy Hodler about vector geometry, graph topology, memory, and why the next generation of AI needs both.
Interview with Roie Schwaber-Cohen.
How sparse and dense retrieval complement one another—and why real search quality needs more than a single representation.
Talk by Roie Schwaber-Cohen at Haystack US 2023.
How domain-specific evaluation and smaller evaluator models can improve reliability while controlling the cost and latency of production AI.
Webinar presented by Roie Schwaber-Cohen and Jim Bennett.