Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabs @aicouncilconf
Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabs  @aicouncilconf
Uploaded June 2026 | Updated November 2026, 2 days ago
[2026 - DAY 3 - LIGHTNING TALK] Video streams combine vision, audio, time-series and semantics at a scale and complexity unlike text alone. At TwelveLabs, we’ve found that tackling this challenge doesn’t start with ever-bigger models — it starts with engineering the right context. In this session, we’ll walk engineers and infrastructure leads through how to build production-grade video AI by systematically designing what information the model receives, how it's selected, compressed, and isolated. You’ll learn our four pillars of video context engineering (Write, Select, Compress, Isolate), see how our foundation models (Marengo & Pegasus) and agent product (Jockey) use them, and review real-world outcomes in media, public-safety and advertising pipelines. We’ll also dive into how you measure context effectiveness — tokens per minute, retrieval hit rates, versioned context pipelines — and how this insight drives cost, latency and trust improvements. If you’re deploying AI video solutions in the wild, you’ll leave with a blueprint for turning raw video into deployable insight — not by model size alone, but by targeted context engineering.

SPEAKER:
James Le - Head of Developer Experience, TwelveLabs

👉 Sign up for our "No BS" Newsletter to get the latest technical data & AI content: aicouncil.com/newsletter

ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.

FIND US:
Website: aicouncil.com
LinkedIn: linkedin.com/company/aicouncilconf
X: https://x.com/aicouncilconf
Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabsNobody agrees on how to build AI responsibly. Thats why this conversation matters.Prompt to Production: Building from Traditional SaaS to True AI-Native SaaS | WebflowSuper-Secret Next Big Thing for DuckDBMultimodal data: The future of data engineeringVik Korrapati is building vision AI the slow way. Its working.A Modern Data Stack in HealthcareTextQL’s Ethan Ding: Every question has a price. Make sure the answer is worth more.Snowflake’s Bala Kasiviswanathan: Your data doesnt need a new home. It needs context.Models as Tools: My Perspective On the MatterIntroducing Pixeltable: Open Source Data Infrastructure for Multimodal AIA Local-First Approach to Extremely Fast Streaming Visualization
AI Council |

Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabs

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER