From Notebooks to Production FASTERClaimThe team serves about 200 data scientists and develops platform tools with their feedback.2:01
Threads / Serving and deployment / 2026
Serving and deployment in 2026
7 sessions
From Notebooks to Production FASTER
PodcastSpeed and Scale: How Today's AI Datacenters Are Operating Through HypergrowthClaimNetBox is the system of record for infrastructure, modeling everything from space, power, and cooling through racks, GPU servers, switching, cabling, IP addresses, configurations, and automation.1:39
Fast & Asynchronous: Drift Your AI, Not Your GPU BillClaimData scientists use plain Python functions that take dictionaries and return dictionaries, while platform teams manage the deployment and autoscaling configuration.12:01
Why AI Agents Shouldn't Replace Your Fraud ModelsClaimHigh-stakes systems need auditability, high query throughput, and low latency, so they are better suited to rules engines and machine learning models than to full agentic decisioning.6:49
MCP Goes StatelessClaimUnder the older protocol, scaling across MCP server instances can produce session-not-found errors because session data is kept in an individual instance's local memory.5:25
MCP 2.0: What Changes, What Breaks, and What You Need to Do About ItClaimStateless MCP servers are easier to scale because any replica can answer any client request without relying on a sticky session.3:58
MCP Release Overview: Stateless and the Big Changes in the New SpecClaimThe new MCP specification is the biggest protocol change since MCP launched and reflects MCP's shift from local use toward remote client-server deployments.1:11