Skip to content
View FlosMume's full-sized avatar
4 followers · 13 following

Block or report FlosMume

Block user

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
FlosMume/README.md

Samuel Huang — AI/ML Engineer | LLM Systems, RAG, and SFT

I build applied LLM systems end-to-end: ingestion → retrieval (Chroma/FAISS) → orchestration → UX (Streamlit/FastAPI). Recent focus: healthcare decision support, evaluation, and safe deployment.

Featured Projects

  • CareMind-Streamlit — Clinical decision support demo (Streamlit UI over a RAG pipeline). Python, Streamlit, Chroma, SQLite

What I’m good at

RAG pipelines • Prompt/Eval frameworks • qLoRA/Fine-tuning • LangChain/LangGraph • Vector DBs • Python packaging • CI/CD • Cloud deployment

Pinned Loading

  1. CareMind-Streamlit CareMind-Streamlit Public

    End-to-end healthcare RAG pipeline built with Streamlit and ChromaDB — includes LLM-based retrieval, SQLite drug DB, and contextual evidence reasoning.

    Python

  2. AI-Research-Assistant-Starter AI-Research-Assistant-Starter Public

    Prototype of an intelligent research agent capable of literature retrieval, summarization, and contextual reasoning — a foundation for scientific automation tools.

    Python 1

  3. LLM-Safety-Labs-Starter LLM-Safety-Labs-Starter Public

    Foundation for building safer generative-AI systems — includes example safety labs for bias detection, toxicity analysis, and RLHF-based response alignment.

    Python

  4. LLAMA-qLoRA-Unsloth-Starter LLAMA-qLoRA-Unsloth-Starter Public

    Fine-tuning Llama models with QLoRA using Unsloth for supervised instruction tasks

    Python

  5. cpp-cuda-image-filter cpp-cuda-image-filter Public

    CUDA 2D image blur using shared memory tiling and constant memory for efficient convolution on GPU.

    Shell

  6. cpp-cuda-deepvision-rtx-starter cpp-cuda-deepvision-rtx-starter Public

    CUDA C++ practice project for RTX 4070 SUPER — explore GPU concurrency, pinned memory, and Nsight profiling. Includes SAXPY and 2D blur kernels to train optimization, stream overlap, and timing ana…

    Cuda