Ujwala Sankh — AI Researcher

AI/ML professional with 7+ years of experience in Research & Development, specializing in Machine Learning, Deep Learning, NLP, Transformers, Embedded AI, and Model Optimization. Strong expertise in end‑to‑end development of custom models—from dataset preparation and model training to deployment and inference acceleration. Hands‑on experience with model optimization, quantization (down to 2‑bit), pruning, TensorRT, OpenVINO, and hardware‑specific accelerations on Intel VPUs (MyriadX), NVIDIA Jetson platforms (TX2, AGX, Nano), and latest Intel PC architectures (MTL, LNL, PTL, NVL). Previously part of LTTS CTO – Computer Vision R&D team for 4 years, driving innovations in edge vision systems, object detection, and multi‑sensor AI workloads. Currently working at Intel as a Software Development Engineer (Innovation Team), enabling and optimizing cutting‑edge AI and GenAI models on next‑gen client CPUs and NPUs, enhancing performance, throughput, and energy efficiency. Proven track record in: Custom model development from scratch GenAI model optimization & inference acceleration Low‑level optimization for edge devices Quantization & compression for real‑world deployment Cross-platform hardware-aware AI engineering Strong problem solver with a research mindset and experience transforming PoCs into production‑ready, industry‑standard AI solutions.

Stackforce AI infers this person is a highly skilled AI/ML engineer specializing in model optimization and embedded AI solutions.

Location: Bengaluru, Karnataka, India

Experience: 8 yrs 1 mo