Cut Through the Context Hype: 4 Layers Your Agent Is Missing
Cut Through the Context Hype: 4 Layers Your Agent Is Missing
- Day
- Day 3 — Session Day 2
- Time
- 3:45pm-4:05pm
- Room
- Track 8
- Track
- Context Engineering
Accessible with the Engineering pass and above.
About this session
AI without context is confidence without comprehension. Your agent can parse queries, generate SQL, and produce articulate explanations — and still be completely wrong. Not because the model is weak. Because it has no world model of your business. This talk cuts through the context engineering hype and gets concrete. Drawing on 500+ enterprise deployments — Mastercard, Nasdaq, GM, Rocket Mortgage, CME Group — Prukalpa Sankar breaks down the four layers every agent is missing: the data it can access, the meaning behind it, the business rules nobody ever wrote down, and the user context that determines what a correct answer actually looks like. Even OpenAI needed six layers of structured context to make their own internal data agent work — cutting query time from 22 minutes to 90 seconds. The pattern is consistent across every production deployment we've seen. The takeaway: a practical four-layer audit framework you can apply to your own stack on Monday.
