Selected work
Menu Planner
Nutrition planning that makes documents and AI useful without making them the interface.
Planning starts with uneven inputs: documents, recipes and external services do not arrive in a consistent shape. The product needs to turn that material into data a person can inspect and continue editing.
Context
The problem to solve
Unstructured documents and unreliable upstream services create friction at the exact moment a user needs a clear plan.
A product flow that turns document input into editable planning data and degrades safely when automation is unavailable.
Product decisions
Decisions that shape the flow
Keep imported information editable
OCR and AI support data entry; the planning workflow keeps the resulting information visible for a person to review.
Keep a usable path when automation fails
Fallback paths make upstream service failures explicit while preserving the core planning task.
Treat integration boundaries as product behavior
Document input, AI processing, persistence and the interface stay separate so the workflow can recover at each step.
Reviewable evidence
Evidence to inspect
- OCR ingestion
- AI-assisted planning
- Resilient delivery
These points describe implementation scope. They do not claim measured performance or user outcomes.
Technologies
- Nuxt
- Neon
- OpenAI
- Docker