Serverless ML on AWS: How I Cut Inference Costs by 99% with Lambda & ONNX

TL;DR: I built an AutoML platform on AWS that handles training and inference for pennies. By moving...