Free open-source course no-shortcuts: 173 lessons + 166 runnable labs from DE to RL

DeepRatAI · reddit · 2026-10-27

A data practitioner open-sourced BlueRingsLabs/no-shortcuts, a free course with 173 lessons and 166 runnable labs for Python users moving into DE/ML/DL/AI—inspired by a first ML job that was mostly wrangling a 400M-row dirty table. Ten parts cover foundations, data engineering, classical ML, deep learning, language models, AI engineering, RL, MLOps, and capstones. Labs are .py files with %% cells (diff-friendly): build autograd, train a GPT (tokenizer, KV cache), SFT/LoRA/DPO, RAG + eval, prompt-injection break/fix, ZeRO/FSDP math, quantization/pruning/distillation, dynamic batching for p99, DQN/PPO. Runs on an 8GB laptop, no GPU needed.

Original post →

More from Companies & People

Companies & People channel →