-
zombie requests
-
practices i’m carrying into 2026
-
model serving system 101
-
understand distributed model training
-
how to use automatic mixed precision in pytorch for optimizing model training
-
simplify neural nets with pruning and compression
-
build an efficient data pipeline
-
tune pytorch to optimize model training using OpenMP
-
under the hood of a powerful feature - torch.compile
-
intro to torch compile
-
the llama 3 herd of models
-
some basics of docker
-
speculative decoding
-
programming on GPU with openai triton
-
how to enable high performance LLM serving
-
crash course on CUDA
-
how to automate your ml project with gitlab and dataiku
-
setting up a paperspace project from scratch
-
how to use tmux and git as a ml engineer