Visual Layer: From Video Search to Agentic Visual AI
Details
Video is the largest source of data that companies own but it is basically a wasted asset because they can’t search inside it to find what they need!!
Join Visual Layer’s research & engineering teams for a deep dive into the architecture required to turn raw pixel data into searchable, explorable knowledge at the petabyte level.
Agenda:
- Arrival & Networking
- Welcome & Intro
- The State of the Art in Video Temporal Search
- Lessons Learned Building the Visual Layer Chat
- Open Discussion & Q&A
We’ll walk through the engineering decisions behind building our "Visual Layer Chat": turning video frames into a graph-based knowledge system. You'll see how we design agents over visual graphs, handle failure modes in production, and bridge the gap between raw visual data and natural language queries.
Who should attend:
- Engineers and Data Architects
- AI researchers
- Product Owners of natural language chats and search engines
- CPOs, CTOs and VPs building search or retrieval-augmented generation (RAG) on visual data
Speakers:
- Amir Alush CTO
- Amir Baron, Senior Backend Engineer
- Guy Singer, Algo Research Lead
When: February 11, 2026, 18:00-19:45
Where: Visual Layer offices, Tel Aviv. WeWork Toha. 114 Yigal Alon, Tel Aviv, Floor 11.
Registration required Drinks & snacks provided