My Open Source Projects
Here’s a collection of projects I’ve been building. All code is open source on GitHub.
runanywhere-sdks
Production-ready toolkit to run AI locally
A Kotlin SDK designed to make running AI models locally straightforward and production-grade. Built with a focus on ease of use and reliability for on-device AI inference.
AI-on-Arm
AI inference on ARM hardware
Exploring and optimising AI model inference on ARM-based hardware. This project is about understanding how to get the best performance from AI models running on ARM chips — from setup through benchmarking.
Mecanum-Wheel-Robot
ROS-based mecanum wheel robot
A robotics project using ROS (Robot Operating System) to control a mecanum wheel robot. Covers locomotion, motion planning, and sensor integration.
SAM2forAV
SAM2 model applied to autonomous vehicles
Applying Meta’s Segment Anything Model 2 (SAM2) to autonomous vehicle perception tasks. Explores zero-shot segmentation for AV sensor data.
AEB-Model-Based-Design
Automatic Emergency Braking — Model-Based Design
A model-based design implementation of an Automatic Emergency Braking (AEB) system, a safety-critical feature in modern vehicles.
perfetto_analysis_repo
Performance analysis with Perfetto
Python scripts for capturing and analysing system traces using Google’s Perfetto tracing tool. Useful for deep-diving into CPU scheduling, memory, and I/O behaviour.
libfreenect2
Open-source drivers for Kinect for Windows v2
Forked and contributed to the open-source driver for Microsoft Kinect v2, enabling depth camera access on Linux and macOS.
decentralized-voting-using-blockchain
Decentralised voting with fingerprint authentication
A blockchain-based voting system that uses MFS100 fingerprint scanner authentication to ensure secure, verifiable, and tamper-proof elections.
More coming soon. Follow me on GitHub to stay updated.