FDA research systems / applied AI / AWS architecture
What I’ve led and what I’m building next.
I’m leading the first LLM integration into FDA’s Research Tracking System and the planning to replatform RTS to AWS. The work connects today’s research workflow to an eventual modernization path.
01 / Research Tracking System / applied AI
Let AI do the comparison. Let experts make the decision.
I’m leading the first AI integration into FDA’s Research Tracking System (RTS). The in-boundary LLM analyzes regulatory research abstracts against the governing coding manual and returns recommendations, supporting evidence, and discrepancy flags.
This workflow is being integrated into RTS to reduce manual comparison work and direct expert attention to disagreements. Regulatory science staff remain responsible for each decision.
Reduce low-value comparison work
Direct expert attention toward disagreements and edge cases.
Ground every recommendation
Use the governing manual and surface the evidence behind the output.
Keep reviewers accountable
The system recommends. FDA staff decide.
Design for the agency boundary
Integrate into the existing research workflow and security environment.
Plan the move to AWS around the work RTS must support.
I’m leading the planning effort to replatform RTS to AWS and establish a path for eventual modernization. The decisions cover the current system, the target architecture, the migration sequence, and the people who will operate it.
Current system
Understand the research workflows, integrations, data, and constraints that the replatform must preserve.
AWS architecture
Shape a target architecture around security, resilience, access, operations, and cost before selecting services.
Modernization path
Sequence migration and later improvements so RTS continues to serve scientists as the platform changes.
Modernizing the information layer behind regulatory science.
At FDA’s Center for Tobacco Products, I connect research operations, data integration, scientific retrieval, acquisition, and product delivery. Much of the implementation is internal; the mission and several core systems are documented publicly.
CTP Integrated Research Data System
CIRDS
I helped lead CIRDS from proof of concept toward production and am first author on FDA’s public technical poster about the platform. CIRDS integrated more than 15 million records from four previously disconnected sources into a unified search and discovery experience for CTP scientists.
View the FDA technical poster ↗Integrated records
Research projects and documents brought together across four heterogeneous sources.
Source systems
RTS, TCKB, iDAT, and the UCSF Truth Tobacco Industry Documents archive.
Mission context
FDA’s current IT plan describes AIRRS as a CTP Office of Science suite supporting scientific and business analysis.
FDA IT plan ↗One platform, connected decisions.
I direct TutorialRepo across AWS delivery, AI, publishing, a native iOS client, and commercial operation. It is a working example of carrying architecture choices through launch and support.
Reusable foundation
Shared course, practice, review, and progress capabilities support additional subjects without rebuilding the system.
Cloud to client
AWS content and AI delivery connect to the SwiftUI application through a publishing workflow.
Operational ownership
Product and architecture decisions continue through subscriptions, releases, and support.
View the working product ↗