FlosMume
Machine Learning Engineer specializing in Large Language Model systems. Experience with retrieval-augmented generation, QLoRA fine-tuning, and evaluation of sa
Pinned Repositories
CareMind-Streamlit
End-to-end healthcare RAG pipeline built with Streamlit and ChromaDB — includes LLM-based retrieval, SQLite drug DB, and contextual evidence reasoning.
AI-Research-Assistant-Starter
Prototype of an intelligent research agent capable of literature retrieval, summarization, and contextual reasoning — a foundation for scientific automation tools.
LLM-Safety-Labs-Starter
Foundation for building safer generative-AI systems — includes example safety labs for bias detection, toxicity analysis, and RLHF-based response alignment.
LLAMA-qLoRA-Unsloth-Starter
Fine-tuning Llama models with QLoRA using Unsloth for supervised instruction tasks
cpp-cuda-image-filter
CUDA 2D image blur using shared memory tiling and constant memory for efficient convolution on GPU.
cpp-cuda-deepvision-rtx-starter
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 analysis for NVIDIA Developer Technology Engineering skillset.
autonomous-multi-terrain-rover
An autonomous multi-terrain rover built with ESP32 and ESP32-CAM, featuring real-time video streaming, ultrasonic obstacle detection, and PWM-driven motor control. Designed for confined-space navigation, inspection robotics, and embedded systems education.
cpp-cuda-starter
CUDA C/C++ starter template for Windows 11 + WSL2 (RTX 4070 SUPER tested)
cpp-cuda-streams-and-pinned-mem
A CUDA C++ demo showing how to overlap data transfer and kernel execution using multiple streams and pinned (page-locked) host memory. This project illustrates asynchronous memcpy, event timing, and performance benefits of concurrent GPU execution — essential for building high-throughput pipelines.
cpp-cuda-thust-intro
Thrust (CUDA) by example: transform/zip, scan, reduce, sort—minimal C++/CMake samples that run on WSL2/RTX.
FlosMume's Repositories
FlosMume/xgb-starter
FlosMume/autonomous-multi-terrain-rover
An autonomous multi-terrain rover built with ESP32 and ESP32-CAM, featuring real-time video streaming, ultrasonic obstacle detection, and PWM-driven motor control. Designed for confined-space navigation, inspection robotics, and embedded systems education.
FlosMume/credit-approver-classifier
Credit approval classification using Logistic Regression and Decision Tree with scikit-learn (Conda + Jupyter).
FlosMume/CUDA-AI-Inference-Starter
FlosMume/FlosMume
FlosMume/cpp-cuda-deepvision-rtx-starter
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 analysis for NVIDIA Developer Technology Engineering skillset.
FlosMume/cpp-cuda-streams-and-pinned-mem
A CUDA C++ demo showing how to overlap data transfer and kernel execution using multiple streams and pinned (page-locked) host memory. This project illustrates asynchronous memcpy, event timing, and performance benefits of concurrent GPU execution — essential for building high-throughput pipelines.
FlosMume/cpp-cuda-thust-intro
Thrust (CUDA) by example: transform/zip, scan, reduce, sort—minimal C++/CMake samples that run on WSL2/RTX.
FlosMume/cpp-cuda-image-filter
CUDA 2D image blur using shared memory tiling and constant memory for efficient convolution on GPU.
FlosMume/cpp-cuda-starter
CUDA C/C++ starter template for Windows 11 + WSL2 (RTX 4070 SUPER tested)
FlosMume/CareMind-Streamlit
End-to-end healthcare RAG pipeline built with Streamlit and ChromaDB — includes LLM-based retrieval, SQLite drug DB, and contextual evidence reasoning.
FlosMume/LLAMA-qLoRA-Unsloth-Starter
Fine-tuning Llama models with QLoRA using Unsloth for supervised instruction tasks
FlosMume/AI-Research-Assistant-Starter
Prototype of an intelligent research agent capable of literature retrieval, summarization, and contextual reasoning — a foundation for scientific automation tools.
FlosMume/LLM-Safety-Labs-Starter
Foundation for building safer generative-AI systems — includes example safety labs for bias detection, toxicity analysis, and RLHF-based response alignment.
FlosMume/MLE_in_Gen_AI-Course