I am a Senior ML Research Engineer at Qualcomm AI. 🔭
I am currently working on LLM efficiency and their application in different systems. 💬
I previously worked on:
- LLM sparsity and cache efficiency: Efficient LLM Inference using Dynamic Input Pruning and Cache-Aware Masking (MLSys 2025)
- Machine Learning for GNSS localisation: GNSS Positioning using Cost Function Regulated Multilateration and Graph Neural Networks (ION GNSS+ 2023)
- Personalization and test-time adaptation of Face anti-spoofing models: Online Adaptive Personalization for Face Anti-spoofing (ICIP 2022), A Personalized Benchmark for Face Anti-spoofing (WACV 2022 workshops)
- Image-to-Graph Generative models: Image-Conditioned Graph Generation for Road Network Extraction (NeurIPS 2019 workshops)
- Generative vision models for medical imaging: Adding Object Detection Skills to Visual Dialogue Agents (ECCV 2018 workshops)
- Object detection in visual question answering: Chest X-ray Inpainting with Deep Generative Models (2018 arXiv preprint)
The repositories in this GitHub profile are 6+ years old, and mostly cover research and coursework projects from my MSc and BSc. They are therefore not representative of code quality and implementation practices in my daily work. 🤔