# Sahil M. > AI @ NVIDIA Location: San Francisco, California, United States Profile: https://flows.cv/sahilm Interested in training LLMs and agents to speed up everyday tasks! ## Work Experience ### Senior Machine Learning Engineer @ NVIDIA Jan 2024 – Present | San Francisco Bay Area Developing agents and training models for CUDA ### Senior Systems Software Engineer @ NVIDIA Jan 2024 – Jan 2024 | Santa Clara, California, United States ### Deep Learning Software Engineer @ NVIDIA Jan 2022 – Jan 2024 | Santa Clara, California, United States TensorRT ### Graduate Research Assistant @ Gupta AI Lab Jan 2021 – Jan 2022 | Urbana, Illinois, United States I researched deep-learning-based methods for interactive object understanding and object tracking in egocentric videos. I was advised by Assistant Professor Saurabh Gupta. ### Software Engineering Intern (TensorRT) @ NVIDIA Jan 2021 – Jan 2021 | Santa Clara, California, United States • Introduced a software-based kernel timing heuristic (DLSim) for neural network optimization • Bridged C++ TensorRT and Python DLSim by implementing a server-client interface handling 10s of queries/s with convolutional layer parameter translation • Compared 5 networks (ResNet-50, MobileNet, Inception-V4, etc), across 3 batch sizes and FP16, INT8 precision, meeting the <10% throughout reduction goal ### Software Development Engineer Intern @ Amazon Jan 2020 – Jan 2020 | Seattle, Washington, United States • Worked in a 6 person team to improve frontend asset generation • Improved Javascript asset generation build time by 18.5% and decreased memory usage by 11% • Analyzed code syntax trees for unfavorable behavior, decreasing final asset size by 5% • Designed a variant generation algorithm an order of magnitude faster for server built variants and client responsive variants ### Undergraduate Research Assistant @ Distrubuted Autonomous Systems Laboratory Jan 2020 – Jan 2020 | Urbana, Illinois, United States • Investigated vision-based robot heading estimation with a self-supervised network on PyTorch achieving 2 degrees error • Devised a supervised network for autonomously calculate pose and drive a robot with distance to intervention of 30 meters • Augmented video data with homographic transformations to simulate robot variance and increase dataset coverage ### Computer Vision Research Intern @ EarthSense, Inc. Jan 2019 – Jan 2019 | Champaign, Illinois, United States • Ascertained intrinsic camera matrix of Terrasentia robot cameras • Achieved 92% accuracy for corn ear height estimation from video by fusing a neural network with single view metrology ### Software Engineering Intern @ Northrop Grumman Jan 2019 – Jan 2019 | Rolling Meadows, Illinois • Designed and developed a C# application to configure and test missile warning and deterrence algorithms. • Pulled in development schedule by 2 months. • Implemented an algorithm to determine orientation from a set of points with singular value decomposition. ### Computer Vision and Data Analysis Intern @ EarthSense, Inc. Jan 2018 – Jan 2019 | Champaign, Illinois, United States • Trained a Convolutional Neural Network with TensorFlow on a 90-10 split dataset that classifies the lodging of wheat with 80% accuracy. • Deployed a ML model to count plant stems with 96% accuracy • Proposed additional data collection metrics to address overfitting ### Perception Engineering Intern @ Swarm Robotix Jan 2018 – Jan 2018 | Naperville, IL • Designed software architecture for an autonomous swarm of MACJac robots • Wrote vision algorithms with OpenCV that detected and tracked corner castings • Researched Convolutional Neural Networks to improve vision accuracy • Implemented SLAM with A* path planning for real-time navigation in the environment ## Education ### Master of Science - MS in Computer Science University of Illinois Urbana-Champaign ### Bachelor's degree in Computer Science University of Illinois Urbana-Champaign ### Engineering Academy Bartlett High School ## Contact & Social - LinkedIn: https://linkedin.com/in/sahil-modi - Portfolio: https://sahilmodi.dev --- Source: https://flows.cv/sahilm JSON Resume: https://flows.cv/sahilm/resume.json Last updated: 2026-03-29