Build Automated Data Pipeline + AI Daily Summary (Dutchie POS → Google Sheets → WhatsApp/Email)

I run a cannabis dispensary using Dutchie POS. I want a SIMPLE and RELIABLE automated system that pulls daily data, analyzes it, and sends me a short actionable summary.

This is a backend automation project. No dashboards or UI.

----------------------
PHASE 1 (REQUIRED)
----------------------

1. Data Extraction
- Pull daily data from Dutchie POS:
- Sales by product (daily totals)
- Inventory snapshot (current stock levels)
- Customer data

- API preferred if available
- If API is not available, use structured CSV automation (not manual)

2. Data Storage
- Send data into Google Sheets
- Must be clean and structured (no messy formatting)

Tabs required:
- Sales (date, product, quantity, revenue)
- Inventory (product, stock level)
- Customers (name, total spend, last visit if available)

3. Automation Workflow
- Use Zapier, Make, or similar
- Fully automated, runs daily without manual input

4. AI Summary (OpenAI)
- Generate a SHORT, actionable summary:

Output format:
- Revenue: $
- Top 3 products
- Slow / no-selling products
- Low inventory items
- 1–2 recommended actions

(No long paragraphs)

5. Notifications
- Send summary daily to:
- Email
- WhatsApp (Twilio or WhatsApp API required)

----------------------
IMPORTANT
----------------------
- No dashboards
- No front-end apps
- Must be simple, stable, and easy to maintain
- Clean structure is critical

----------------------
PHASE 2 (OPTIONAL)
----------------------
- Real-time alerts (low inventory, no sales periods)
- Customer segmentation + automated SMS marketing

----------------------
REQUIREMENTS
----------------------
- Experience with APIs or data extraction
- Experience with Zapier or Make
- Experience integrating OpenAI
- Experience with Twilio (or WhatsApp API)

----------------------
IN YOUR PROPOSAL
----------------------
Please include:

1. How you will pull data from Dutchie (API or CSV approach)
2. Tools you will use
3. Estimated timeline
4. Examples of similar automation/data pipeline work

This may lead to ongoing work if Phase 1 is successful.

Thanks