Skip to content

Projects

Things I have built, mostly from first principles, mostly to understand them.

  1. 01 · Reinforcement learningRL TestbedAn open laboratory for learning and doing RL: 29 algorithms from tabular Q-learning to world models and GRPO, implemented readably, with live training, seed comparisons and a learning path.PythonPyTorchReact
  2. 02 · World modelsSurgical Vision World ModelA world model that learns surgical scene dynamics from unlabeled video. MICCAI 2025 workshop.PyTorch
  3. 03 · NLPExtractive Nepali QAMuRIL fine-tuned for extractive question answering in Nepali, a low-resource language.PythonTransformers
  4. 04 · Language designGandivAn interpreter for a custom programming language: lexer, parser and runtime, from scratch.C++
  5. 05 · Game enginesApollo / DungeonRushA 2D engine on an Entity–Component–System architecture, used to build a complete platformer.C++ECS
saurav@kathmandu: ~esc to close