What if AI wasn't a chatbot added to a business application, but part of the business itself?
For Hacktoberfest 2026, I decided to build something around a problem that is actually happening in my community.
I built Collectify, an AI-first inventory and business management platform designed for a real local store that sells anime and pop-culture collectibles such as Funko, Pokémon products, plushies, pins, posters, and other merchandise.
This is not a hypothetical business scenario.
The project is being developed around the needs of a real local store with multiple partners, where keeping track of products, ownership, sales, inventory, earnings, and purchasing decisions can become difficult when information is handled manually or spread across different places.
And that is where the idea for Collectify started.
The Real Problem
A traditional inventory system can tell you:
"You have 5 units of this product."
But a real business needs to ask much more:
- Who owns those products?
- Which partner's products are selling?
- Which products are performing better?
- What should be restocked?
- What products are becoming slow-moving?
- How much is each partner generating?
- Where could we find suppliers?
- What should we consider purchasing next?
- What does the business data actually tell us? For a store with multiple partners, inventory is not just about counting products. It is also about understanding the business behind the inventory. That is the problem Collectify is designed to address. Built for a Real Business Collectify is being developed around a real local store, not an imaginary customer created specifically for the hackathon. The store has multiple partners who contribute and sell products through the same physical location. That creates a very practical challenge: How do you maintain control over everyone's products and sales while still having a unified view of the business? Collectify addresses this by connecting: Products → Partners → Sales → Earnings → Inventory → AI The objective is to give the people running the store a centralized place where they can understand what is happening and make better decisions.
The Idea Behind Collectify
I wanted to avoid a very common pattern in AI applications:
Inventory System
+
Generic Chatbot
=
"AI-powered" application
That was not the direction I wanted to take.
Instead, Collectify follows this model:
Business Data
↓
AI
↓
Understanding
↓
Recommendations
↓
Business Action
The AI is not there simply to answer questions.
It is designed to interact with the information generated by the business and turn that information into something useful.
Why AI Is Actually Useful Here
Collectify uses AI in workflows that have a direct relationship with the business.
For example:
Inventory Data
↓
AI Analysis
↓
Restocking Insights
Sales Data
↓
AI Analysis
↓
Product Performance
Partner Data
↓
AI Analysis
↓
Individual Business Insights
Product Search
↓
AI Supplier Discovery
↓
Potential Suppliers
↓
Purchasing Decision
The important part is that the AI is connected to the application's actual domain.
It is not simply sitting beside the application as a chat window.
AI Business Analysis
Collectify includes an AI analysis workflow where users can ask questions about their business.
For example:
"Which products are performing better and which ones should I consider restocking?"
The goal is to transform operational information into business insight.
Instead of manually going through multiple records, the user can interact with the business data through the AI layer.
The long-term idea is to make Collectify capable of continuously helping the business understand:
- Product performance
- Sales behavior
- Inventory conditions
- Partner performance
- Potential restocking needs
- Business opportunities
AI-Powered Supplier Discovery
One of the most important AI workflows in Collectify is supplier discovery.
A user can search for something like:
"Funko Pokémon"
Collectify sends the request through the backend AI workflow and returns structured supplier information.
Instead of returning only a paragraph of text, the system can produce application-ready information such as:
- Supplier name
- Description
- Category
- Location
- Website
- Relevance
- Rating when available
- Verification information when available
- Supplier tags The frontend then consumes those results and renders them as supplier cards. This distinction is important. The AI is not only generating language. It is generating structured information that the application can use. User ↓ Supplier Search ↓ FastAPI ↓ AI / Supplier Discovery ↓ Structured Results ↓ Angular ↓ Supplier Cards
AI Is a Business Tool, Not a Decoration
This is probably the most important design decision behind Collectify.
I did not want to be able to say:
"My inventory application has AI."
I wanted to be able to say:
"The AI changes what the application can do."
Without the AI layer, Collectify can store business information.
With the AI layer, the system can begin to interpret that information and help transform it into decisions.
That creates a much more interesting relationship:
Store
↓
Data
↓
Understanding
↓
Decision
↓
Action
This is the direction I want to continue developing.
Multi-Partner Management
Another important part of Collectify is the multi-partner model.
The store is shared by multiple partners, which means the system needs to understand that not every product belongs to the same person.
Collectify allows the business to organize information around individual partners and their products.
This helps provide visibility into:
- Product ownership
- Individual inventory
- Sales associated with products
- Partner information
- Earnings
- Business performance Instead of having one large inventory with no context, Collectify creates a more structured representation of the business.
Inventory Management
Collectify provides the operational foundation required by the AI layer.
Products can contain information such as:
- Name
- Description
- Category
- SKU
- Cost
- Sale price
- Stock
- Status
- Ownership The inventory is not intended to be the final product. It is the data foundation that allows the rest of the platform to work.
Sales and Earnings
Sales are connected to products and business information so that the platform can later analyze what is happening.
The earnings section provides another layer of visibility into the operation.
These sections are important because they provide the data that future AI workflows can use for deeper business analysis.
Open AI and Local Development
One of the requirements of this Hacktoberfest challenge is to build something with open-source AI at its core.
Collectify was designed around an AI layer that can work with open/open-weight models and different providers, rather than making the entire application permanently dependent on a single proprietary model.
The application was also developed and tested locally, including both the Angular frontend and the FastAPI backend.
This is important to me because I want the AI layer to remain replaceable.
Collectify
↓
AI Service
↓
Model / Provider
The application should be able to evolve as models evolve.
That means being able to:
- Experiment with different models.
- Replace providers.
- Test different AI capabilities.
- Adapt the AI layer without redesigning the entire application.
- Run the core application locally. An AI-first application should not necessarily mean being locked into one model forever. Evidence of the AI Workflow The screenshots below document the actual development and functionality of the AI features.
Technology Stack
Frontend
- Angular
- TypeScript
- SCSS
- Angular HttpClient
- Responsive UI Backend
- Python
- FastAPI
- Pydantic
- Uvicorn
- REST API Database
- Supabase
- PostgreSQL AI
- OpenRouter
- Open/open-weight AI models
- AI-powered business analysis
- Structured AI supplier discovery Deployment
- Vercel
- Render
- Supabase Architecture ┌─────────────────────────┐ │ Collectify │ │ Angular Frontend │ └────────────┬────────────┘ │ HTTP/REST │ ▼ ┌─────────────────────────┐ │ FastAPI Backend │ │ │ │ Products │ │ Sales │ │ Partners │ │ Earnings │ │ AI Analysis │ │ Supplier Discovery │ └────────────┬────────────┘ │ ┌──────────┴──────────┐ │ │ ▼ ▼ ┌──────────────────┐ ┌──────────────────┐ │ Supabase │ │ OpenRouter │ │ PostgreSQL │ │ AI Models │ └──────────────────┘ └──────────────────┘
Running Locally
Clone the repository:
git clone https://github.com/Ezequie1Sc/collectify.git
Install the frontend dependencies:
cd frontend/collectify
pnpm install
Run the frontend:
pnpm start
The Angular application will be available at:
http://localhost:4200
For the backend:
uvicorn app.main:app --reload
The API will run at:
http://127.0.0.1:8000
Health check:
http://127.0.0.1:8000/health
The project was developed and tested locally before deployment.
Live Project
Frontend
https://collectify-7xcu.vercel.app
Backend API
https://collectify-api-udxk.onrender.com
API Documentation
https://collectify-api-udxk.onrender.com/docs
GitHub Repository
https://github.com/Ezequie1Sc/collectify
Why This Project Matters
Small businesses generate valuable information every day.
But that information is often distributed across:
- Spreadsheets
- Messages
- Notebooks
- Manual records
- Different applications
- Conversations between partners The problem is not necessarily the lack of data. The problem is the lack of understanding. Collectify attempts to create a bridge between operational data and business decisions. Inventory ↓ Sales ↓ Business Data ↓ AI ↓ Insights ↓ Recommendations ↓ Action
That is the direction I believe AI can bring real value to small businesses.
What Makes Collectify Different?
Collectify is not trying to compete with large enterprise ERP systems.
The goal is different.
It is about taking a real small-business problem and asking:
What could happen if the business's own data became something the application could understand?
The project combines:
Inventory management
Multi-partner management
Sales and earnings
AI analysis
AI-powered supplier discovery
An AI-first business platform
And the most important part is that this architecture was shaped by an actual business problem.
What's Next?
Collectify is still evolving.
Some of the features I want to explore next include:
- Sales forecasting
- Automatic restock recommendations
- Partner-specific analytics
- Supplier comparison
- Purchase recommendations
- Profitability analysis
- Automated business insights
- Role-based access control
- Advanced dashboards
- More open-weight AI models
- More autonomous AI workflows The long-term goal is to move from: Inventory Management
towards:
AI-Powered Business Intelligence
for Small Retail Businesses
Final Thoughts
This project changed the way I think about AI applications.
AI does not have to be a chat window placed next to an existing application.
It can become part of the application's actual workflow.
With Collectify, the goal is:
Organize Information
↓
Understand the Business
↓
Discover Opportunities
↓
Make Better Decisions
And most importantly, this is not a hypothetical problem.
Collectify was built around a real local business, with real operational needs and a real multi-partner inventory problem.
That made the project much more meaningful to build.
I did not start with:
"What AI feature can I add to my application?"
I started with:
"What problem does this business actually have, and where can AI genuinely help?"
That question shaped the entire project.
AI should not be decoration. It should solve a real problem.
Built for Hacktoberfest 2026
Hacktoberfest Weekend Challenge — Build for a Friend
Collectify — AI-powered business management for real-world retail.
GitHub:
https://github.com/Ezequie1Sc/collectify
Live Demo:
https://collectify-7xcu.vercel.app
API Documentation:
https://collectify-api-udxk.onrender.com/docs
Top comments (0)