I Build Things.
Let's chat! Drop me a text! Email LinkedIn
I enjoy tackling interdiciplinary problems. Thriving at the intersection of mechanical systems, electronics, software development and system design, I always try to approach problems holistically.
Whether it’s iterating on aerodynamic designs, building AI agents, or crafting intuitive user experiences, I leverage my diverse skills to build efficient, impactful, and scalable solutions.
What I've Done!
DJI RoboMaster DART team
Collaborated with NUS Calibur Robotics under Undergrad Research Experience to develop a high accuracy projectile launching robot to base future iterations on.
Building Agentic AI systems
Spent a summer fighting AI frameworks, pipelines and endless documentation. As unpredictable and frustrating they can be, engineering a reliable system from the ground up was a fun experience, so... here's a concoction of my thoughts from building an agentic orchestration system.
Autonomous Heat-Seeking Bot
Built a fully autonomous robot using ROS2, capable of navigating a maze, detecting thermal sources, and launching projectiles. The system integrated pathfinding and heatsensing together with mechanical flywheels.
Implementation of Fundemental ML Concepts
Built increasingly complex ML models ranging from simple bigram models to a basic transformer model. During which, implemented concepts from the following papers; Wavenet a generative model for raw audio, Attention is all you need, Deep residual learning for image recognition, Improving neural networks by preventing co-adaptation of feature detectors ML Paper (Dropout Technique)
Soybean Decision Support System: Leveraging NASA Satellite Data
NASA SpaceApps 2024, My team developed a data-driven approach towards optimizing decision making in selecting appropriate Soy Beans by regions
