Cover Image for AI Infrastructure Night
Cover Image for AI Infrastructure Night
Avatar for Alien
Presented by
Alien
Ship to your customer's cloud
439 Went
Register to See Address
San Francisco, CA
Registration
Registration Closed
This event is not currently taking registrations. You may contact the host or subscribe to receive updates.
About Event

​​About the event

​AI Infrastructure Night brings together founders and engineers building the systems behind production AI.

​Four technical talks on building, deploying, and operating production AI systems.


​✨ Speaker lineup

​5:00–6:00 PM — Arrival, Food & Networking

​Lotte Verheyden, Developer Relations Engineer, Langfuse
Running & Maintaining AI Agents in Production
What changes once an agent moves beyond a prototype? A technical look at observing, evaluating, debugging, and maintaining agents in production.

​John Kupchanko, Developer Relations Engineer, Qdrant
Can 1,000 Flies Share One Memory?

​Researchers mapped every neuron in a fruit fly's brain, which got me thinking: what if artificial brains could learn from each other's mistakes? I built an experiment with 1,000 simulated flies, each in its own world. Using Qdrant as shared memory, they could learn from what other flies experienced. The results were pretty amazing. Flies sharing memory made the right call 95% of the time, compared to 50% alone. 999 survived with shared memory, compared to just 82 alone, and the entire fleet only sent 1.5 MB to the cloud.

​Alien, Aditya Mahamuni, Ambassador @ Alien
How Do You Operate Software Inside Infrastructure You Don’t Control?
Deploying into a customer’s cloud once is straightforward; upgrading and debugging it across hundreds of opinionated environments is not. This talk covers drift, permissions, partial failures, upgrades, rollbacks, observability, and supporting Terraform, Helm, and internal platform tooling without direct access.

​Nick Bushak, Co-founder, Boxes
AI Coding in the Cloud with Boxes.dev
Most developers still run coding agents like Claude Code and Codex locally, but parallel agents can quickly introduce challenges around compute, isolation, and testing. This talk explores what changes when development environments move to the cloud, using one Linux machine per agent. We’ll look at the architecture behind this approach, the tradeoffs compared with local development, and what we’ve learned running coding agents in cloud environments.

​7:00–8:00 PM — Networking

Location
Please register to see the exact location of this event.
San Francisco, CA
Avatar for Alien
Presented by
Alien
Ship to your customer's cloud
439 Went