Skip to content
View dokutoshi's full-sized avatar
12 followers · 42 following

Achievements

Achievement: QuickdrawAchievement: Pull Sharkx3

Achievements

Achievement: QuickdrawAchievement: Pull Sharkx3

Block or report dokutoshi

Block user
See More

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
dokutoshi/README.md

0→1 applied scientist and researcher, with a background spanning edge ML hardware (FPGA-based inference devices) through modern LLM/RL systems. 26 US patents and applications.

Facilitates the Reinforcement Learning and Reasoning Models tracks for a Data Science & Machine Learning collaborative learning group, working through Sutton & Barto (SAB), the Farama stack (Gymnasium, PettingZoo), and Raschka's Reasoning Models from Scratch.

Repositories
nvidia-rag RAG over 59,016 pages of NVIDIA documentation. Compares vector stores (Chroma, FAISS), embeddings (ada-002, all-mpnet-base-v2) and chunk sizes; RAGAS evaluation with a self-hosted judge; 8-bit and GGUF quantization for local inference.
hf-llm-course HuggingFace LLM course material — tokenization, fine-tuning, causal LM pre-training, SFT and post-training.
rl-gym RL fundamentals — Blackjack (Monte Carlo vs TD), CartPole, a custom GridWorld environment with a Gymnasium wrapper, and Taxi-v3 action masking.

Interested in reasoning, evaluation, and the gap between research code and production systems.

Pinned Loading

  1. hf-llm-course hf-llm-course Public

    HuggingFace LLM course material — tokenization, fine-tuning, causal LM pre-training, SFT and post-training. Facilitated for the Data Science & Machine Learning Collaborative Learning Group (~7,000 …

    Jupyter Notebook

  2. LLMs-from-scratch LLMs-from-scratch Public

    Forked from rasbt/LLMs-from-scratch

    Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

    Jupyter Notebook

  3. reasoning-from-scratch reasoning-from-scratch Public

    Forked from rasbt/reasoning-from-scratch

    Implement a reasoning LLM in PyTorch from scratch, step by step

    Jupyter Notebook

  4. rl-gym rl-gym Public

    Reinforcement Learning: Gymnasium/PettingZoo (Farama), Sutton & Barto, HuggingFace RL — session material for the DS/ML Collaborative Learning Group

    Python

  5. anthropics/claude-code anthropics/claude-code Public

    Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

    Python 143k 22.9k

  6. nvidia-rag nvidia-rag Public

    RAG over 59,016 pages of NVIDIA documentation — Chroma vs FAISS, ada-002 vs all-mpnet-base-v2, RAGAS evaluation with a self-hosted judge, 8-bit and GGUF quantization for local inference

    Jupyter Notebook