This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle
What I Built
I built SellerSpark AI, a privacy-first AI product listing generator for small online sellers.
Writing product listings can be time-consuming, especially for small businesses and individual sellers. SellerSpark AI allows users to enter simple product details and generates a complete, marketplace-ready listing.
It generates:
- 🏷️ Product titles
- 📝 Product descriptions
- ✨ Selling points
- 🔍 Search keywords
- #️⃣ Hashtags
The target users are small online sellers, marketplace sellers, handmade-product businesses, and social-media sellers who want to create professional product listings faster.
A potential customer would pay for this because it can save time and reduce the effort required to create optimized product content for multiple products.
Demo
🌐 Live Demo: https://ayush-kasaudhan-01.github.io/SellerSpark-AI/
The basic workflow is:
Enter Product Details → Generate Listing → Copy & Use
The AI runs directly in the browser, so users don't need to set up a backend or provide an API key.
Code
💻 GitHub Repository: https://github.com/ayush-kasaudhan-01/SellerSpark-AI.git
The complete source code is available in the repository.
How I Built It
SellerSpark AI is built around an open-weight AI model running directly in the browser.
AI Stack
- SmolLM2 — open-weight language model
- Transformers.js — browser-based AI inference
- ONNX — model execution
- JavaScript — application logic
- HTML & CSS — frontend and user interface
The application takes the seller's product information and creates a structured prompt for the open-weight model.
The model then generates the product title, description, selling points, keywords, and hashtags.
One of the main goals was to avoid depending on a traditional closed AI API. The project works as a browser-based AI application without requiring a backend server or API key.
Why Does Open Innovation Matter?
Open innovation made it possible for me to build an AI-powered side-hustle product without relying completely on a paid closed AI API.
Using an open-weight model provides:
- 🔒 Better privacy
- 💰 Lower operating costs
- 🔄 Freedom to experiment with different models
- 🧩 More control over the AI workflow
- 🚫 Less vendor lock-in
- 🌐 The possibility of running inference directly in the browser
For small businesses and independent developers, these advantages can make AI tools more accessible and affordable.
SellerSpark AI is an example of how open AI technology can be turned into a practical product that solves a real business problem.
Prize Categories
- Side Hustle
- Open-Source AI
Top comments (0)