Vision Stock

• 

  InvyIQ

---

InvyIQ – Intelligent Inventory Automation from Final Product Sales

The Origin Story

Inventory systems today focus heavily on manual tracking of raw materials. But for restaurants, bakeries, and food trucks, this approach breaks down fast. Staff are too busy to count every ingredient; spoilage and missed orders become common. That’s when we asked:

What if we could reverse-engineer inventory from what was sold instead of what was stocked?

That’s how InvyIQ was born — an intelligent system that automatically deducts ingredient usage by simply logging final products sold.

---

What We Built

InvyIQ uses a product mapping model: Each final product (e.g., “Chicken Sandwich”) has a defined bill of ingredients. When a sale is logged, InvyIQ deducts the associated raw materials — no counting needed.

Key features:

• 21-day AI forecasting based on actual usage
• Inventory alerts based on thresholds
• CSV import/export
• Smart dashboard in Bolt (.tsx)
• Auto-generated reorder lists

---

The AI Leap: Automating Vendor Calls

Early on, we had one pain point: Even after generating reorder lists, a human still had to call vendors — one at a time.

This didn’t scale.

So we introduced AI-powered outbound calls using Retell AI:

• All vendors get called automatically via a single webhook.
• Calls include dynamic order details (e.g., “Please send 20 lbs of chicken breast…”).
• If vendors don’t answer, AI leaves a voicemail.
• Simultaneously, order confirmation emails are sent and logged in Google Sheets.

Result: Managers click one button — and AI takes care of the rest.

---

The Challenges We Faced

• Webhook chaining: Coordinating order generation → vendor-specific call scripts → tracking outcomes.
• Batching logic: Initially, each vendor had to be clicked manually. We upgraded to a single batch-trigger.

• Speech script handling: Passing dynamic variables ({{order_list}}, {{vendor_name}}) into Retell AI required testing and retry logic.

• Testing limitations: Using the same phone number for vendor testing caused voicemails, but it worked — meaning true scalability is possible.

---

What’s Next for InvyIQ?

After the hackathon, we plan to:

• Build a vendor dashboard with real-time order tracking
• Add Slack/SMS notifications for low inventory
• Enable AI forecasting per ingredient (not just per final product)
• Integrate with POS systems for automated sales imports
• Offer multi-location support for franchises

We’re excited to bring real AI automation into the daily grind of small food businesses — saving time, reducing errors, and preventing spoiled or missed orders.

---

Why It Matters

InvyIQ bridges the gap between sales and supply — automatically. In a world where small businesses are under pressure to do more with less, this kind of intelligent automation can make or break success.

---

Built With

• bolt.new
• javascript
• make.com
• retellai
• roboflow
• supabase