Why it matters
Talks on foundation-model acceleration and orchestration infrastructure speak directly to the compute and inference-efficiency tradeoffs that determine real-world AI deployment costs.
The tokenmaxxing angle
Sage Elliott's talk on AI orchestration with Flyte and Union.ai covers the exact pipeline layer where model routing and compute-cost decisions get made for production ML systems.
From the organizers
The agenda names NVIDIA Research's Bowen Wen presenting FoundationStereo, Voxel51's Prerna Dhareshwar on FiftyOne, and Union.ai's Sage Elliott on Flyte orchestration, hosted at Union.ai's Bellevue office.