Prarthana Bhattacharyya

I’m a Machine Learning Research Engineer at Eedi, specializing in building practical ML solutions that bridge cutting-edge research and real-world applications. With 7+ years of experience spanning autonomous driving, wearable AI, and education technology, I focus on developing efficient, deployable models that solve complex problems.

My research has been recognized with a Best Paper award at ECCV-2024 Smart Eyewear Workshop and cited 450+ times, with open-source contributions achieving 400+ GitHub stars. My work has been demonstrated at major industry events including AWE-USA and CES.

I’m a UK Global Talent awardee and hold a Ph.D. in Electrical and Computer Engineering from the University of Waterloo and a Master’s from the University of Tokyo. I speak English, Bengali and Hindi fluently, with N2 certification in Japanese proficiency.

Current Focus

At Eedi, I develop models that improve personalized learning at scale. My work combines educational and generative AI with practical MLOps to deliver measurable impact for students and educators.

Research Interests

Event-Based Vision & Wearables: Pioneered ultra-low power gesture recognition systems for always-on smart eyewear, achieving state-of-the-art accuracy at only 6mW power consumption.
Autonomous Driving: Developed learning approaches for motion forecasting and 3D object detection, enabling robust perception in complex urban environments.
Efficient ML: Specializing in building high-performance models with practical constraints: low power, limited data, real-time deployment. My ML work has been shipped in production for AI music tutor applications.

Technical Expertise

Deep learning frameworks (PyTorch, TensorFlow, Keras), MLOps platforms (AWS, Azure, MLFlow, ClearML), 3D computer vision, self-supervised learning, graph neural networks, and event-based vision.

selected publications

Uncertainty-Aware Knowledge Tracing Models

Joshua Mitton^*, Prarthana Bhattacharyya^*, Ralph Abboud, and 1 more author

Pre-print, Under review, 2025

TL;DR arXiv

New uncertainty-based approach that identifies struggling students where traditional metrics fail, enabling EdTech platforms to intervene at critical moments.
Helios 2.0: A Robust, Ultra-Low Power Gesture Recognition System Optimised for Event-Sensor based Wearables

Prarthana Bhattacharyya^*, Joshua Mitton^*, Ryan Page^*, and 8 more authors

Pre-print, Under review, 2025

TL;DR arXiv PDF Video

Architected mobile-optimized models for smart glasses achieving 25x power reduction with 20% accuracy gain (F1>80%). Profiled inference bottlenecks, implemented quantization and efficient architectures. Deployed at CES-2025 in commercial development kit.
Helios: An extremely low power event-based gesture recognition for always-on smart eyewear

Prarthana Bhattacharyya^*, Joshua Mitton^*, Ryan Page^*, and 8 more authors

Best Paper, Proceedings of the European Conference on Computer Vision (ECCV) Workshops, 2024

TL;DR arXiv PDF Video

Pioneered low-power low-latency ML models, representations and custom loss functions enabling micro-gesture detection for Meta-Raybans. Rapidly delivered a prototype within 6 months, publicly demonstrated at AWE-USA 2024 and featured by CNET. Won Best Paper award at ECCV-2024 Workshop.
SSL-Lanes: Self-Supervised Learning for Motion Forecasting in Autonomous Driving

Prarthana Bhattacharyya, Chengjie Huang, and Krzysztof Czarnecki

In Conference on Robot Learning (CoRL), 2022

TL;DR arXiv PDF Code

Pioneered SSL methods for autonomous vehicle motion forecasting (80+ citations, 120+ GitHub stars, presented to Gatik AI). Achieves state-of-the-art prediction accuracy while maintaining lower computational complexity than transformer-based approaches.
Sa-Det3D: Self-Attention Based Context-Aware 3D Object Detection

Prarthana Bhattacharyya, Chengjie Huang, and Krzysztof Czarnecki

In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops , 2021

TL;DR arXiv PDF Video Code

First method to combine self-attention with CNNs for 3D point cloud detection (90+ citations, 150+ GitHub stars). 1.5 AP accuracy gain while slashing model size by 15-80% and compute by 30-50%. Proven across KITTI, nuScenes, and Waymo datasets.
Fantrack: 3d multi-object tracking with feature association network

Erkan Baser, Venkateshwaran Balasubramanian^*, Prarthana Bhattacharyya^*, and 1 more author

In 2019 IEEE Intelligent Vehicles Symposium (IV), 2019

TL;DR arXiv PDF Code

NeurIPS 2019 3D tracking competition winner (150+ citations). Replaces hand-crafted cost functions and complex optimization with a CNN that learns end-to-end. Competitive multi-object tracking by automatically learning optimal assignments.