A Machine Learning Framework for Large-Scale Behavioral Monitoring to Assess Thermotolerance in Dairy Cows
IEEE Access
Hi, I'm
ML Software Engineer
Building intelligent systems at the intersection of machine learning and software engineering.

I'm a Software Engineer and a Research Scientist at Prairie View A&M University, where I develop computer vision systems for agricultural AI applications. My work spans from training YOLO models for behavioral monitoring to building speech recognition systems for low-resource African languages. I'm passionate about applying machine learning to solve real-world problems and making AI more accessible.
Prairie View A&M University
GPA: 3.66 | 2025 - Present
Focus: Computer Vision, Deep Learning
Babcock University
2024
Baze University
GPA: 3.5 | 2021
Prairie View A&M University
Debyl Ltd / Shell Corporation
Huawei Technologies
Architected a pipeline processing 38,000+ videos to quantify drinking and brushing behaviors in dairy cattle, achieving 96% classification accuracy to aid heat stress research.
Developed the first ASR system for the low-resource Ika language by fine-tuning Meta's MMS and Wav2Vec2 models, including a custom forced-alignment data pipeline.
A multi-camera computer vision system using homographic projection and color-based re-identification to track individual birds in high-density poultry environments.
An edge-computing framework correlating environmental gas emissions with poultry activity levels to analyze the impact of Spirulina-supplemented diets.
Serverless streaming ASR system on AWS supporting 5 African languages with hot-swappable language adapters. Achieved 4x inference speedup through INT8 quantization.
Open-source toolkit for livestock monitoring featuring pre-trained models for poultry detection and tools for multi-animal pose estimation dataset creation.
Chess engine trained from scratch using self-play and Monte Carlo Tree Search with custom ResNet backbone. Capable of defeating intermediate-level human opponents.
Custom linear algebra library implemented from scratch to visualize memory overhead and optimization techniques compared to standard Python lists.
Optimized extension using custom CUDA kernels to accelerate matrix operations by 100x on NVIDIA GPUs.
Real-time data ingestion pipeline for a cryogenic plant, handling pressure and thermal sensor data with sub-second latency for engineering diagnostics.
Graph database analysis of character interactions in Brandon Sanderson's Cosmere using entity extraction to map hidden connections between worlds.
Fantasy Premier League optimization tool that fetches real-time API data to predict player points and recommend squad transfers.
Custom 2D physics and rendering engine built from scratch in C++ to simulate frictionless movement mechanics.
IEEE Access
arXiv
Under Review
Under Review
I'm always interested in hearing about new opportunities, collaborations, or just having a chat about ML and software engineering.