Neel Mokaria

Software Engineer & AI/ML Researcher.

Hit me up
Scroll Down

Hi, I'm Neel

Software Engineer & AI/ML Researcher at UMD. I build systems that matter - from knowledge graphs to award-winning ML pipelines - and push the boundaries of what's possible with AI.

I do things

I combine sophisticated technologies to make something useful(sometimes not).

Computer Science student at the University of Maryland, College Park (graduating Dec. 2026), working at the intersection of systems engineering and AI research. My stack spans TypeScript, Python, React, PostgreSQL, AWS, and more - from architecting dual-protocol APIs to building multimodal AI pipelines and publishing ML research.

Beyond the keyboard, I'm an avid swimmer, squash player, and music nerd. Endlessly curious - whether it's exploring distributed systems, NLP, or the latest in agentic AI.

Work Experience

Hack4Impact-National

Software Engineer - Volunteer Management System Aug. 2025 – Present

For All Ages × Hack4Impact-UMD

Software Engineer Sep. 2025 – Present

MIND Labs, UMIACS @ University of Maryland

Research Intern - Knowledge Graph, NLP & Agentic AI May 2025 – Present

GAMMA Labs, UMIACS @ University of Maryland

Research Intern - Multimodal Agentic AI Jun. 2025 – Feb. 2026

Division of Information Technology, UMD

Classroom Support: Field Tech Team Feb. 2024 – Present

Research & Recognition

Publications & Awards.

Publication

Foundations and Frontiers of Multimodal Agentic Frameworks

Gamma Labs, UMIACS @ UMD - under Prof. Dinesh Manocha  ·  Jun. 2025 – Feb. 2026

Surveyed and evaluated multimodal agentic frameworks to surface open research gaps by analyzing reasoning capabilities and cross-modal information integration across benchmark performance, scalability, and applicability to real-world multimodal AI tasks. Submitted to Transactions on Machine Learning Research (TMLR); under review.

1st Place

Ironsite Spatial Intelligence Hackathon, UMD

First Place Hackathon Winner  ·  Feb. 2026

Built a three-stage ML pipeline (YOLO11 → SAM3 → LoRA-tuned Qwen3-VL-8B with 3-pass adversarial verification) to automatically detect PPE non-compliance, restricted zone breaches, and proximity hazards from low-res dual-camera construction footage, mapping violations to OSHA standards with severity scoring. Built a FastAPI backend and Next.js dashboard with appearance embeddings for persistent worker tracking and automated end-of-shift safety reports.

Expertise

Skills & Projects

Python JavaScript TypeScript Java C/C++ SQL HTML5 CSS3 React NodeJS Firebase MongoDB Flask Django Docker AWS PostgreSQL PyTorch Git GraphQL Python JavaScript TypeScript Java C/C++ SQL HTML5 CSS3 React NodeJS Firebase MongoDB Flask Django Docker AWS PostgreSQL PyTorch Git GraphQL

I make things

Featured Projects

Hit me up!

I'd love to talk.

Always looking for opportunities to work on hard problems - whether in research, engineering, or at the intersection of both. Open to internships, research collaborations, and full-time roles. Let's build something great.

Find me on the web

Loading...