Selected Projects
Four projects across reinforcement learning, network security, cloud-native infrastructure, and simulation. Click any card to expand deep architectural details.
Tetris Deep RL Research
ResearchA controlled study of why Tetris agents learn from hand-crafted features and fail completely on the raw board.
Work Experience
Engineering roles where I build computer vision pipelines, perception models, and real-time robotic systems.
Computer Vision Engineer
Perception for a humanoid robot: finding the ball and reading the field from a live camera feed.
- ›Ball detection: RF-DETR (Large), fine-tuned through Roboflow, reaches 97.7% mAP@50 with 97.1% precision and 97.0% recall on a held-out split of a 2,684-image multi-ball dataset. A YOLO detector was trained alongside it.
- ›Field segmentation: YOLO segmentation trained on 10,233 annotated images across venues and lighting runs at about 12 FPS, and was benchmarked against RF-DETR Small.
Technical Skills
A comprehensive overview of programming languages, deep learning frameworks, security tooling, and infrastructure technologies I work with.
Programming & ML
Digital Forensics & Security
Infra & Networking
Certifications
AWS Certified AI Practitioner
Validates foundational AI/ML concepts, generative AI models, prompt engineering, and secure deployment on AWS cloud infrastructure.
Amazon Web Services
Verify on Credly ↗
Cloud Web Application Builder
Validates designing, building, and deploying resilient, scalable cloud-native web architectures utilizing AWS compute, VPC, and storage.
AWS Academy
Verify on Credly ↗
CCNA: Introduction to Networks
Validates foundational networking architectures, IPv4/IPv6 variable-length subnetting, Ethernet switching, routing, and network security.
Cisco Networking Academy
Verify on Credly ↗
Transformer Models with PyTorch
Explores self-attention mechanisms, multi-head attention, positional encodings, encoder-decoder architectures, and fine-tuning transformer models.
DataCamp
Statement #47,642,753 ↗
Deep Learning for Images with PyTorch
Mastery in Convolutional Neural Networks (CNNs), feature extraction, pooling, transfer learning with pre-trained models, and image recognition.
DataCamp
Statement #47,589,282 ↗
Intermediate Deep Learning with PyTorch
Hands-on training in custom PyTorch training loops, loss functions, optimizers, activation functions, and deep multi-layer neural architectures.
DataCamp
Statement #47,585,866 ↗