From 15% to 90% GPU Utilization: Fix the Data Pipeline, Not the Model
An AI Engineer talk on taking GPU utilization from 15% to 90% by fixing the data pipeline instead of touching the model.
More in Infra
About this video
AI Engineer · October 10, 2026 · 17:02
AI EngineerGPUdata pipelinetalk
Open on YouTube ↗More in Infra
More Infra
When Running Your Own AI Models Beats Paying for APIs - Steven Latré, imec
Serving World Models: Latency vs Throughput with NVIDIA Cosmos 3 Super on Nebius | Cosmos Labs
Why Humans Can't Keep Up With AI-Written Code - Tariq Shaukat, Sonar
