Siddharth Mehta
- AI / ML
- Data Science
- ML System Optimization
- Computer Vision
- MLOps
- Full-Stack Systems
Hi, I'm Siddharth Mehta — a Machine Learning Engineer and Master's student at the University at Buffalo. I work across applied ML, large-scale data science, and full-stack engineering, building production-ready applications. My experience spans research, data science, and software development, and I thrive at the intersection of algorithms, systems, and real-world impact. Explore my projects or read my articles to get a deeper look at my work, ideas, and engineering approach.
Project Highlights
View More
A tiny scalar-valued autograd engine and a neural net on top of it. A C++ version of the karpathy/micrograd repo.
- AI
- ML
- Machine Learning
- + more
Experience
University at Buffalo
Research Assistant
Aug 2025 - Dec 2025
Developed quantitative sampling-coverage metrics and optimized physics-based imaging systems through 3D simulation, vectorized PyTorch ray-tracing (85% speedup), and Bayesian optimization for large configuration search, enabling scalable experimental workflows and contributing to publications at SPIE Imaging, UNYAPM, and IEEE TMI.
Visit
Walmart Global Tech
Data Science Intern
Jun 2025 - Aug 2025
Built and deployed large-scale ML/LLM systems for content understanding, distribution-shift monitoring, automated evaluation, and personalized feature ranking, extracting high-signal insights from thousands of artifacts while reducing manual effort by 95% and delivering autoscaled real-time performance on GCP.
Visit
ZS Associates
Data Science Associate
Jun 2022 - Jul 2024
Developed large-scale PySpark/Hadoop pipelines and XGBoost-based prioritization models to power product analytics and next-best-engagement decisions, reducing rep workload by 20%, improving targeting accuracy, and driving $100K+ incremental revenue through data-driven customer-level actions.
visit
Lakebrains Technologies
Machine Learning Intern
May 2020 - Jul 2020
Engineered real-time edge-compute systems combining face recognition with motion-detection–based activity recognition, optimized and deployed on Raspberry Pi and Jetson Nano for efficient, low-latency performance.
National Institute of Technology Hamirpur
Research Assistant — Biometric Signal Analysis
Aug 2020 - Jun 2022
Conducted research on ECG and PPG signal processing for non-invasive blood pressure estimation, implementing hybrid neural network models and validating performance on the MIMIC-II dataset.
VisitNational Institute of Technology Hamirpur
Training & Placement Representative
Aug 2021 - Jun 2022
Coordinated with companies and students to streamline the on-campus recruitment process, managed data for candidate shortlisting, and supported logistics for placement drives.
Visit
Rhythmeecz dance club - National Institute of Technology Hamirpur
Dance Club Secretary
Aug 2021 - Jun 2022
Led the official dance club, organized cultural events and performances, managed team operations, and represented the institute in inter-college competitions.
Visit