Moss Live Lab: Ship Sub-10ms Retrieval in Under an Hour

​Most retrieval stacks add 200-500ms of network latency to every query. That kills real-time voice agents, copilots, and any app where users are waiting.

​Moss runs retrieval in-process. No network round-trip. Sub-10ms.

​In this live session, you'll:

• ​Set up Moss and build a semantic index from scratch

• ​Run real queries and see latency numbers live

• ​Walk away with working retrieval you can plug into your own app

​This is hands-on. You'll be writing code alongside us.

​Before the session: Create your free account at Portal and complete onboarding (2 clicks). Bring your own data or pick one of the ready-made indexes on the portal.

​Open to all levels. If you're building AI agents, this is worth an hour.