FIRST-PRINCIPLES AI · RESEARCH TO DEPLOYMENT

I build AI systems that see, reason, and act in the real world.

I build AI systems where technical depth meets real-world consequence. I have taken models from research into applied settings, including real-time edge inference on portable devices and contributions to an FDA-cleared diagnostic product. My focus is general AI and computer vision: representation learning, multimodal sensing, edge optimization, and robust decision loops under constraint.

Start a conversation
Shubham

Current focus

Now

Designing real-time AI systems that remain reliable under strict latency and hardware constraints.

Core strengths

Representation learning, computer vision, edge optimization, data strategy, and end-to-end model ownership.

Direction

Building general AI systems that combine sensing, reasoning, and action in high-stakes environments.

Industry experience

2025 – Now
Data and AI Engineer/Philips

Architecting scalable AI infrastructure, model-serving systems, and reliable deployment pipelines.

2020 – 2025
AI Research Scientist/Philips

Built real-time edge AI for detection, tracking, and classification, and translated research into deployable clinical systems.

2019 – 2020
Research Assistant/Rochester Institute of Technology

Developed deep learning pipelines for aerial perception and multispectral segmentation.

2019 – 2019
R&D Intern/Philips

Built early 3D analysis and synthetic-data pipelines that later scaled into broader AI programs.

Selected papers

View all research →