Projects

Here are some projects I have worked on.

F3AMD: Fast Autoregressive Motion Diffusion

F3AMD: Fast Autoregressive Motion Diffusion

SCA 2026 / Computer Graphics Forum. Co-authored research on efficient multi-character motion synthesis with Fourier Neural Operators, FiLM conditioning, and low-step diffusion sampling.

Learn more →
Digital Twin - Your Data, Your Rules

Digital Twin - Your Data, Your Rules

A conversational AI agent that maintains structured memory of user-specific facts, mimics user face and voice, and exposes memory as an interactive surface for data governance. This project combines neuroscience-inspired memory architecture with multimodal embodiment and interactive memory governance.

Learn more →
Learned Motion Matching

Learned Motion Matching

An implementation of Learned Motion Matching with deep neural networks for real-time character animation. This project combines traditional motion matching algorithms with deep learning to create smooth, responsive character animations.

Learn more →
Reinforcement Learning Soccer Bot

Reinforcement Learning Soccer Bot

Training intelligent soccer-playing agents using reinforcement learning algorithms in DeepMind Soccer Environment. This project implements and compares SAC, PPO, and POCA algorithms to develop autonomous multi-agent soccer players.

Learn more →
Total Recall

Total Recall

A Software Framework for Capturing Events of Interest from Long Surveillance Videos

Learn more →
Single Shot Detection

Single Shot Detection

This article implements Single Shot Detection (SSD), a fast object detection model using multi-scale feature maps and anchor boxes. It follows the standard SSD pipeline with a VGG16 backbone, enabling real-time performance with efficient bounding box predictions.

Learn more →
Restaurant Quality App

Restaurant Quality App

An Android application to show qualities of restaurants in Surrey, Vancouver.

Learn more →