Building Gastric IQ: Digestion Intelligence for the GLP-1 Journey

Navigating life on GLP-1 medications (like Mounjaro, Ozempic, or Wegovy) or recovering from bariatric surgery requires a completely different approach to nutrition. Standard fitness apps are built around a simple calories-in, calories-out philosophy. But when your digestive system's mechanics have fundamentally changed, you need more than a food diary—you need context.

I built Gastric IQ to solve this exact problem. It is designed around digestion intelligence, answering the immediate questions users have:

• How full is my stomach right now?
• Is it safe to eat a heavy meal before bed?
• Why are my cravings spiking today?

The Core Differentiators

When scoping this project, the focus was on driving measurable value for users by providing insights that generic trackers miss entirely.

Model-Estimated Gastric Load

Instead of just logging what was eaten, the app calculates a live 0–100% fullness gauge. It factors in meal load, meal type, time elapsed, and the specific GLP-1 effect to estimate current digestion status.

Pharmacokinetic Context

By logging specific medications (tirzepatide, weekly semaglutide, or oral semaglutide), the app maps out the drug's effect over the injection cycle—letting users know if their current medication effect is mild, moderate, or strong.

The Food Noise Index

"Food noise" (intrusive thoughts about food) is a major factor for GLP-1 users. Gastric IQ tracks these cravings and contextualizes them against the user's medication cycle so they know what to expect.

Symptom Correlation

Nausea, reflux, and bloating are stored directly alongside gastric-load and GLP-1 snapshots, making it easy to identify actionable patterns to discuss with clinicians.

Engineering for Scale and Empathy

Building a health-adjacent SaaS product means prioritizing production-readiness, robust API design, and strict data security from day one. Gastric IQ utilizes a secure, multi-tenant architecture designed to scale efficiently while protecting user data.

The system relies on an event-driven flow to recalculate the gastric load model dynamically as new meals, symptoms, or medications are logged.

To reduce friction, I integrated an AI-powered meal recognition system that turns natural-language inputs into structured, editable food entries—complete with protein estimates and their specific impact on gastric load.

On the mobile side, Android Health Connect is integrated to seamlessly pull in weight and body-composition trends, anchoring the app as a central hub for metabolic progress.

Roadmap

The roadmap is expanding heavily into post-bariatric needs—focusing on smaller meal timing, pouch sensitivity, and protein-first recovery journeys.