Software 3.0 and the AI engineer with Swyx
PodRocket
00:00
Discussion on Stochastic Gradient Descent, Putting Foundation Models into Production, and Infrastructure Serving
This chapter discusses the concept of stochastic gradient descent in machine learning and the challenges associated with putting foundation models into production, including context limits, privacy, and rate limits. It also emphasizes the importance of understanding GPU infrastructure, model swap utilization, and the ability to batch for efficient model serving.
Play episode from 08:31
Transcript
