System / 01Oware Engine and a Ladder of RL Agents
2026A custom Oware engine and evaluation ladder for DQN, PPO, AlphaZero-lite, and Minimax agents, including the training strategies that failed.
Hi. I’m Prince.
I’m a systems engineer and ai/robotics researcher. I build autonomous systems and study the places where machine learning, software engineering, and design meet.
This site is my working archive: research notes, open-source systems, long-form essays, and manuscripts shaped by the questions I’m currently following.
At the moment, I’m teaching a practical course on robotics and continuing work across robot learning, perception, and infrastructure for intelligent systems.
01 / Journal
Recent writing
Jul 6, 2026Rigid Body MotionsRoboticsMay 17, 2026Building an Oware Engine and a Ladder of RL AgentsMachine LearningMay 7, 2026A Hybrid Computer Vision Pipeline for Real-Time Traffic Density EstimationMachine LearningApr 6, 2026Configuration Space of RobotsRoboticsMar 23, 2026Dynamic Traffic Shaping by Reinforcement Learning AgentsMachine LearningFeb 1, 2026You have a lot of potentialScience
02 / Research
Recent manuscripts
2026Seamless Deployment of Machine Learning Inference at the Far Edge: An Empirical AWS IoT Greengrass Architecture for Industrial Predictive MaintenanceEdge Computing · IIoT2026Real-Time Traffic Density Estimation: A Hybrid Computer Vision and Predictive Machine Learning PipelineComputer Vision2026Adaptive Network Traffic Shaping with Deep Reinforcement Learning: A PPO-Driven Token Bucket on ns-3Reinforcement Learning · Networking2026The Evolution of xDSL Technologies: Broadband Demands, Technical Principles, and Copper LimitationsNetworking2024A Comprehensive Survey of Side-Channel Attacks in IoT Devices: Techniques and CountermeasuresIoT Security
03 / Systems
Selected work
System / 01
System / 02Real-Time Traffic Density Estimation
2026A two-stage pipeline for traffic monitoring. The first stage pairs an object detector with a spatial logic layer that filters detections against a precomputed road mask, so only on-road vehicles contribute to density measurements.
System / 03Humanoid Balancing Policy with MjLab
2025A stabilizing control policy for the 29-DOF Unitree G1 humanoid trained with PPO and a small Actor-Critic network (512 → 256 → 128). Built on the mjlab framework with MuJoCo Warp to simulate 4,096 parallel environments directly on the GPU, bypassing CPU bottlenecks.