Building AI for Healthcare & Life Sciences

I build AI systems that improve patient care and help organizations figure out where to start.

I’m a Senior AI Solutions Architect at AWS, working with healthcare providers, pharma, MedTech, and payers on the Global Healthcare & Life Sciences team. My job is part engineering, part strategy: figuring out where AI actually moves the needle in healthcare, then building it.

I started in applied math and material science at École des Ponts ParisTech where the curriculum was heavy on continuum mechanics, numerical modeling, and PDEs. No CS degree, but it turns out that learning to model physical systems is a better foundation for AI than most people think. At Columbia, I did an M.S. in Management Science & Engineering, a joint program between the engineering school and the business school - half optimization and analytics, half strategy. That combination shaped how I work: I think in models, but I make decisions in business terms.

I broke into tech through consulting, but what I actually wanted was to build things with real impact. So I joined a medical device company and started putting AI into clinical products: one of the projects led to a US patent. I’ve been in healthcare AI since, delivering ML platforms at AWS Professional Services, leading a GenAI practice at an AWS partner, and now shaping AI strategy for some of the largest life sciences organizations in the world as part of the AWS Healthcare & Life Sciences team.

I speak English 🇺🇸, German 🇩🇪, French 🇫🇷, and Spanish 🇪🇸

Based in New York City.

Point of View

Everyone has access to the same public models. The real advantage is private data: how you structure it, how you extract signal from multimodal sources, and who gets access to what. (Data infrastructure & governance, FM training)
The other bottleneck is inference cost. Custom silicon (I spend a lot of time on AWS Neuron chips and NKI kernels), model fine-tuning, and rigorous agentic evaluation are what determine whether healthcare AI scales.

Publications & Patents

Human-in-the-loop constructs for agentic workflows in Healthcare & Life Sciences — AWS Blog
A guide to building AI agents in GxP environments — AWS Blog
Build a biomedical research agent with Biomni tools and Amazon Bedrock AgentCore — AWS Blog
Accelerate digital pathology slide annotation workflows on AWS — AWS Blog
Move SageMaker Autopilot ML models from experimentation to production using SageMaker Pipelines — AWS Blog
System and Method for Monitoring Clinical Activities — US Patent

Speaking

CitiusTech 2026, Atlantic City — AI for Healthcare & Life Sciences: Building for Breakthroughs (keynote)
AWS Life Sciences Symposium 2026, NYC — Multi-Agents with A2A & MCP on AgentCore (workshop)
AWS re:Invent 2025, Las Vegas — AI and HCP Engagement: Powering Smarter Life Sciences Field Operations
AWS Life Sciences Symposium 2025, NYC — End-to-end productivity for data & AI teams with Amazon SageMaker Unified Studio
AWS New York Summit 2025 — Building Intelligent AI Agents for Life Sciences Innovation
AstraZeneca 2024 — Preparing for an Agentic Future, Global Ethics & Compliance Executive Panel
SBI2 2024, Boston — Transforming Digital Pathology through the Convergence of Cloud and Generative AI

Select Projects

AI Digital Pathology on AWS — open-source framework for digital pathology workflows
Eversana × TensorIoT — GenAI-powered medical regulatory review
Gilead Sciences — ML solutions for pharmaceutical R&D
Moderna — AI platform for commercialization
Masimo VCAM — Visual Clinical Activity Monitoring (AI medical device)
Masimo Sepsis Index — Clinical decision support for sepsis detection

Experience

Amazon Web Services — Sr. AI Solutions Architect, Global Healthcare & Life Sciences (current)
TensorIoT (AWS Partner) — AI/ML Tech Lead
Amazon Web Services — Data Science Lead, Professional Services
Masimo — AI Engineer, Medical Devices
Capgemini — Data Scientist
Air Liquide — Project Engineer, Hydrogen Fuel Cells
Airbus Helicopters / Sogeti — Material Science Engineer

Education

M.S. Management Science & Engineering — Columbia University
Diplôme d’ingénieur, Applied Mathematics & Material Sciences — École des Ponts ParisTech
Classe préparatoire (CPGE) — Lycée Sainte-Geneviève