Ted EdmundsContact

Project detail

FDA Device Intelligence Platform

A production-quality public-data product built around FDA AccessGUDID releases, Python ingestion, SQLite analytics, an Express API, and a dashboard experience.

Case study

Business Problem and Product Vision

Business Problem

FDA AccessGUDID is a rich public dataset, but the raw release archives are not packaged as a decision-support product. Supply chain, manufacturers, regulatory teams, and analysts need faster ways to understand listing velocity, manufacturer activity, product-code momentum, and recently published devices without manually parsing delimited files.

Product Vision

The product vision is a repeatable intelligence platform: download public FDA release ZIPs, preserve raw files for traceability, import delimited tables with Python, build SQLite analytics tables, expose Express API endpoints, and present executives and analysts with a dashboard for search, filtering, trend analysis, and impact-score review.

Users

Product Users

Supply Chain
Manufacturers
Regulatory
Analysts
Executive leadership

Live product

Interactive Dashboard

All confidential data is excluded. Portfolio demos use public data or fictional, portfolio-safe sample data to show KPI hierarchy, trust signals, filtering, and regulated-enterprise product workflows.

portfolio product demo / dashboard

FDA AccessGUDID intelligence

Medical device market intelligence platform

API-first dashboard for public FDA AccessGUDID data. In local mode it consumes the Express API backed by SQLite; on Vercel it falls back to a portfolio-safe snapshot that mirrors the same product model.

Checking local API

Total device listings

5,152,799

SQLite analytics table

Manufacturers

184,276

Distinct company names

New listings this week

11,761

+2.3% WoW

High-impact listings

438

Directional product impact score

Risk/watchlist items

27

Recent listings needing review

Trend

PrimaryTeal = avg impact
Jun 02
Jun 09
Jun 16
Jun 23
Jun 30
Jul 06

Manufacturer activity ranking

Apex BioSystems

+18.4%

1,842 total listings / 241 recent

Northstar Surgical

+11.7%

1,326 total listings / 169 recent

Pulse River Medical

+9.2%

1,108 total listings / 97 recent

Harbor Health Devices

+6.4%

884 total listings / 74 recent

Meridian HealthTech

+5.8%

742 total listings / 62 recent

Portfolio review queue

Supply chain and market intelligence teams can use the impact score to triage high-velocity categories before reviewing individual listings.

Manufacturer momentum

Recently active manufacturers help leaders understand where portfolio, contracting, or competitive monitoring should focus.

Governed public data product

Raw FDA archives remain traceable while analytics tables and API endpoints support repeatable dashboard consumption.

Listings explorer

Search by device name, manufacturer, catalog number, or DI. Sort and page through API-compatible records.

DeviceManufacturerCatalog / DICode / ClassPublish dateStatusImpact

CardioSense Patch XT

Ambulatory ECG monitor

Apex BioSystems

CS-XT-400

00819876543210

DPS

Class II

2026-07-06In Commercial Distribution94.2

OrthoNav Precision Guide

Surgical navigation system

Northstar Surgical

ON-PG-210

00345678901234

HWT

Class II

2026-07-05In Commercial Distribution91.5

GlucoWave Mini Sensor

Glucose monitoring system

Apex BioSystems

GW-M-108

00678901234567

QBJ

Class II

2026-07-04In Commercial Distribution88.9

SonoFlow Vascular Catheter

Cardiovascular catheter

Pulse River Medical

SF-V-88

00112233445566

DQY

Class III

2026-07-03In Commercial Distribution86.7
Page 1 of 2

Recently flagged listings

CardioSense Patch XT

94.2

Apex BioSystems

High listing velocity in cardiac monitoring

NeuroMap Micro Array

84.1

Northstar Surgical

Class III product with sparse category history

SonoFlow Vascular Catheter

86.7

Pulse River Medical

Recent Class III vascular listing

Pipeline status

Last refresh

2026-07-06 07:50 ET

Database size

824 MB local SQLite artifact

Latest release

AccessGUDID weekly release 2026-07-06

Pipeline health

Healthy

Run dateFilesRecordsScoredStatus
2026-07-06315,152,799699,830Completed
2026-06-29295,141,038688,412Completed
2026-06-22335,132,904676,905Completed
1

FDA AccessGUDID

Public delimited release archive from FDA/NLM.

2

Weekly ZIP

Full and weekly releases preserved in raw storage.

3

Python import

Pipe-delimited files loaded and normalized.

4

SQLite analytics

Devices, counts, product baselines, impact scores.

5

Express API

Dashboard endpoints for filters and analytics.

6

Next dashboard

Portfolio surface with API-first live demo.

Architecture

Architecture

Reference architecture

Executive view of source, ingestion, analytics, governance, and dashboard delivery.

Public data / governed pipeline

01

Source systems

FDA AccessGUDID delimited release archive

02

Ingestion

Python download and raw ZIP storage

03

Raw / staging

SQLite raw and staging tables

04

Quality / transform

Analytics tables, quality checks, impact scores

05

Product layer

Express REST API for dashboard queries

06

Experience

React / Next.js portfolio dashboard

GitHub Actions weekly refresh
Governance and traceability
Portfolio dashboard and API consumers

Technical Stack

Data model, pipeline, and implementation

AccessGUDID release archives are downloaded into local raw ZIP storage, imported by a Python pipeline into SQLite, normalized into device and analytics tables, scored for directional market impact, and served through an Express API. The portfolio dashboard is API-first and uses a safe static snapshot when the local API is unavailable on Vercel.

Technology stack

FDA AccessGUDIDPython ETLSQLite analytics databaseExpress REST APIReact / Next.jsTypeScriptTailwind CSSGitHub Actions

Data Product Decisions

Governance, UX, and architecture choices

Treat public FDA files as a governed data product by preserving raw ZIPs and rebuilding analytics tables repeatably.
Separate ingestion, analytics modeling, API access, and dashboard UX so each layer has a clear product responsibility.
Use directional impact scoring to create an analyst review queue while avoiding regulatory or commercial claims.
Design for both local API-backed exploration and Vercel-safe portfolio presentation.

Execution Notes

Challenges solved, lessons learned, and future enhancements

Challenges Solved

  • Parsing large public release archives while keeping raw data traceable and replayable.
  • Normalizing inconsistent delimited source tables into dashboard-ready SQLite analytics views.
  • Designing an API and dashboard that expose useful product signals without overclaiming what public data can prove.

Lessons Learned

  • Public datasets still need product management: users need questions, workflows, trust signals, and refresh expectations.
  • SQLite can be a strong analytics layer for local product demos when paired with clean API boundaries.
  • Dashboard UX should make the pipeline understandable, not hide data freshness and lineage from users.

Future Enhancements

  • Add recall, 510(k), PMA, De Novo, MAUDE, and adverse-event joins for stronger signal context.
  • Publish scheduled GitHub Actions artifacts and API snapshots for repeatable portfolio demos.
  • Add saved views, manufacturer profiles, and product-code alerting for analyst workflows.

What This Demonstrates

What this demonstrates to hiring managers

This project demonstrates data product management, ETL pipeline design, analytics engineering, API design, dashboard UX, data modeling, GitHub Actions automation, and enterprise architecture using a real public FDA data source.

Interested in this kind of data product leadership?

Connect for roles involving enterprise analytics, regulated data platforms, executive dashboards, governance, and AI-ready product strategy.