San Francisco Bay Area
AI researcher and engineer with experience spanning ML Compilers, Inference Optimization, Distributed Systems for Machine Learning, Parallelization, Computer Vision, Medical Image Processing, LLMs, Agentic AI, Memory for Long-Horizon Agents. Currently, I work at AWS (Annapurna Labs, Amazon) for Amazon's Trainium and Inferentia ML Accelerator Compiler Optimizations. I work on the backend of the compiler, primarily in scheduling the low level compute communication instructions to make the best use of the on chip and off chip network and memory. Previously, I worked with DevRev AI on Large-Scale Retrieval, Memory, and Planning for Long-Horizon AI Agents. Before that, at MathWorks, I engineered some of the most requested features for the Sequence Diagram product along with MathWorks' flagship Simulation capabilities, collaborating with SysML’s co-author Alan Moore. At KLA Tencor's Advanced Computing Lab, I focused on accelerating deep learning inference for semiconductor inspection and metrology. I’ve developed novel algorithms for medical imaging, including vessel diameter analysis and explainable AI for Fundus images. My work has been published in Medical Imaging and AI venues. I'm passionate about building robust, scalable AI solutions and advancing the state of the art in Computer Vision and Generative AI. I am looking for SWE/MLE/MTS/Research Engineer full-time roles starting Jan 2027. Please reach out to me at [email protected] for any such opportunities.