AI/ML Engineer based in United States.

I'm Prem, an AI/ML engineer focused on reliable LLM systems, agent workflows, retrieval, evaluation, and applied machine learning. I care about systems that can be measured, explained, and operated under real latency, cost, and quality constraints.

I completed an M.S. in Artificial Intelligence at Rochester Institute of Technology in May 2026 after an official Graduate Researcher appointment on Emotion Engine. Before RIT, I owned a production LLM routing layer at Concentrix + Webhelp serving 50K daily requests across 3 foundation models, cutting p95 latency from 4s to 1.5s and monthly inference spend from $45K to $37K.

My recent public work spans agent infrastructure, search, and product engineering: Kerna wraps agent tool use with fail-closed policy and receipts, Cryo searches a frozen pre-2022 corpus, and Nori and Sushi turn data pipelines into usable products.

The throughline of everything on this site: “Simple systems scale better than clever ones.”

Education

• M.S. in Artificial Intelligence, completed May 2026 · Rochester Institute of Technology, Aug 2024 – May 2026
• B.Tech. in Computer Science, completed May 2024 · National Institute of Technology Silchar, Aug 2020 – May 2024

Experience

• Graduate Researcher, Rochester Institute of Technology · Jan 2026 – Apr 2026
• Generative AI Engineer, Concentrix + Webhelp · Feb 2024 – Jul 2024
• Data Science Intern, AlphaBits Technologies · Aug 2023 – Jan 2024
• ML Engineer Intern, iNeuron AI · Aug 2022 – Aug 2023
• Software Developer Intern, Exposys Data Labs · May 2021 – Jun 2022

Publications

• Lightweight Channel Attention for Efficient CNNs · Designed and evaluated a lightweight channel attention module (LCA) achieving competitive accuracy with negligible parameter and latency overhead on ResNet-18 and MobileNetV2.

Certifications

• Machine Learning Specialization · Stanford Online · Coursera