From curl -fsSL https://mlsysbook.ai/tinytorch/install.sh | bash to a working environment: the real installer runs, then tito setup and Module 01 is built, tested, and exported.
Completing Module 06 (Autograd Engine) with its unit and integration tests passing, then the full 20/20 module status and all seven historical milestones.
Training TinyGPT on Q&A concepts, diagnosing generalization gaps with the Overfitting Detective, and streaming conversational responses token by token.
Training on TinyPy code snippets, then completing Python function names the model never saw in training. Each greedy completion is checked by the AST parser; failures are shown alongside passes.
Measuring INT8 quantization, pruning, im2col convolutions, and KV-cache memoization with pass gates on YOUR modules, and marking the Pareto frontier computed from those measurements.
Interactive terminal companion launcher, quick reference guide, and automated environment diagnostic health checks.
Competitive student benchmarking arena, event scoring, and hardware optimization leaderboards via the Tito CLI.