NutriPrompt is an AI-powered nutrition intelligence platform that transforms structured health data into explainable, contextualized and actionable nutrition workflows.
Built with Django, Prompt Engineering, Retrieval-Augmented Generation (RAG), OCR pipelines, compatibility engines and multi-provider AI orchestration.
NutriPrompt is not a simple meal-plan generator.
It is a production-oriented AI architecture designed to demonstrate how modern intelligent systems can combine structured user data, domain knowledge, retrieval systems, validation layers and resilient orchestration to deliver reliable and explainable outputs.
Nutrition planning is not just a content generation problem.
A reliable AI nutrition system must understand:
- Personal goals
- Dietary restrictions
- Digestive conditions
- Ingredient compatibility
- Nutritional context
- Budget constraints
- Lifestyle habits
NutriPrompt approaches this challenge as an intelligent decision-support workflow, not as a chatbot.
Its architecture enriches every request before generation.
This produces:
✅ Grounded outputs ✅ Fewer hallucinations ✅ Explainable recommendations ✅ Compatibility-aware suggestions ✅ Resilient provider orchestration ✅ Production-oriented AI workflows
Most AI nutrition tools rely on direct prompting.
This creates major limitations:
- No domain grounding
- Weak restriction handling
- No ingredient validation
- No compatibility analysis
- No retrieval logic
- No fallback resilience
For healthcare-adjacent workflows, this is not enough.
NutriPrompt addresses this by structuring AI generation as a layered architecture.
It is not built to simply generate meal plans.
It is built to demonstrate how production-grade AI systems should be designed.
NutriPrompt combines multiple intelligence layers:
- Prompt Engineering
- Retrieval-Augmented Generation (RAG)
- OCR-based document analysis
- Nutrition knowledge systems
- Compatibility engines
- Multi-provider orchestration
- Explainability layers
Instead of sending raw input directly into an LLM, NutriPrompt enriches every request through multiple reasoning layers before generation.
This improves:
- Contextual consistency
- Reliability
- Explainability
- Resilience
User Input
↓
Structured Forms
↓
Profile Analysis
↓
RAG Knowledge Retrieval
↓
Prompt Builder
↓
Gemini API
↓ (Fallback)
OpenAI API
↓
Structured JSON Output
↓
Nutrition Rules Engine
↓
Compatibility Analysis
↓
HTML Rendering
↓
Shopping Intelligence
↓
PDF Generation
↓
AI Copilot Layer
NutriPrompt transforms structured user information into personalized nutrition plans using:
- Goal understanding
- Food preferences
- Digestive symptoms
- Dietary restrictions
- Budget-aware logic
- Real-life context
The AI output includes:
- Structured weekly meal planning
- Compatibility-aware recommendations
- Practical tupper adaptation
- Explainable profile tags
- Downloadable PDF
NutriPrompt transforms the generated plan into an organized shopping list:
- Grouped by category
- Optimized for planning
- Structured for execution
- Practical for real-life use
This turns AI generation into actionable utility.
NutriPrompt also includes a dedicated Streamlit demo built to expose the internal AI workflow in a recruiter-friendly and architecture-focused format.
Designed for:
- Technical presentations
- AI product demos
- Architecture walkthroughs
- Portfolio storytelling
Structured nutrition intake.
End-to-end orchestration pipeline.
Ingredient extraction and incompatibility detection.
Weekly plan generation.
Workflow observability and stack overview.
NutriPrompt includes an explainability assistant capable of:
- Explaining generated plans
- Reviewing restrictions
- Validating ingredients
- Providing contextual reasoning
This transforms NutriPrompt from a generator into an interactive AI decision-support system.
Generate structured weekly plans based on:
- User objectives
- Dietary restrictions
- Symptoms
- Food preferences
- Activity levels
- Budget
- Daily routines
NutriPrompt retrieves nutrition rules before generation.
Examples:
- Low FODMAP recommendations
- Gluten-free alternatives
- Lactose-free substitutions
- Digestive-safe planning
- Shopping optimization logic
This improves consistency and reduces hallucinations.
NutriPrompt analyzes:
- Product labels
- Nutrition PDFs
- Pantry inventories
- Fridge scans
- Ingredient lists
OCR results are validated against user restrictions.
NutriPrompt evaluates:
- User restrictions
- Retrieved nutrition rules
- OCR-detected ingredients
To detect incompatibilities before recommendation.
Generated plans are converted into categorized shopping lists.
This creates:
- Better planning
- Easier execution
- Lower user friction
NutriPrompt implements fault-tolerant provider logic:
Gemini API
↓
OpenAI Fallback
↓
Structured Mock Generation
Benefits:
- Stable demos
- Graceful degradation
- Provider independence
- Predictable outputs
This project showcases:
- Prompt Engineering
- Retrieval-Augmented Generation (RAG)
- OCR Pipelines
- Multi-provider orchestration
- Structured AI outputs
- Explainable AI workflows
- Rule-based reasoning
- Compatibility engines
- Shopping intelligence
- Product-oriented AI architecture
- Service-oriented design
- Fallback resilience
| Layer | Technology |
|---|---|
| Backend | Django |
| Language | Python 3.13 |
| AI Providers | Gemini API + OpenAI API |
| Retrieval | Custom Nutrition RAG |
| OCR | Tesseract OCR |
| Data | JSON |
| PDF Rendering | WeasyPrint |
| Frontend | HTML + CSS |
| Demo Layer | Streamlit |
| Testing | Django Test Framework |
| Architecture | Service-Oriented Design |
NutriPrompt/
├── nutriprompt_app/
│ ├── services/
│ │ ├── ai/
│ │ ├── nutrition/
│ │ ├── profiles/
│ │ ├── rag/
│ │ ├── vision/
│ │ └── presentation/
│ ├── templates/
│ ├── tests/
│ └── views.py
│
├── scripts/
│ └── nutriprompt_demo.py
│
├── docs/
│ ├── screenshots/
│ │ ├── 01_home.png
│ │ ├── 02_plan.png
│ │ └── 03_shopping_list.png
│ └── streamlit_demo/
Current automated validation includes:
- Prompt generation
- Knowledge retrieval
- Compatibility analysis
- OCR processing
- Context injection
- Structured outputs
- Shopping list generation
Run tests:
python manage.py testCurrent suite:
17 automated tests passing
git clone https://github.com/beatriangu/NutriPrompt.git
cd NutriPrompt
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txtpython manage.py runserverOpen:
http://127.0.0.1:8000/
streamlit run ./scripts/nutriprompt_demo.pyNutriPrompt provides informational guidance only.
It does not replace professional medical or nutritional advice.
All outputs should be reviewed by qualified professionals when appropriate.
Bea Lamiquiz
🌐 Portfolio: https://bchill.net 💻 GitHub: https://github.com/beatriangu 💼 LinkedIn: https://www.linkedin.com/in/bealamiquiz/
If you find this project interesting:
⭐ Star the repository 🤝 Connect on LinkedIn 💬 Share feedback, ideas or improvements