This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle.
What I Built
DukaanFlow AI is a local-first AI order desk designed for small tiffin services and home-food businesses.
These businesses often receive customer requests through WhatsApp messages and phone calls. Managing daily changes manually can lead to mistakes, especially when a customer wants to skip one meal but keep their monthly subscription active.
DukaanFlow AI explores how AI can help sellers interpret requests such as:
- "Please skip tomorrow's lunch."
- "I will be away for three days."
- "Cancel my order for Friday."
The goal is to help turn natural-language messages into structured order actions while keeping the seller in control. Important changes should be reviewed before they are applied.
Who would pay for it?
Independent tiffin providers, home-based food sellers, and small meal-subscription businesses could benefit from a tool that reduces manual order-management work. Pricing would need to be validated with real users.
Demo
Live demo: https://utkarshydv27.github.io/DukaanFlow-AI/
The website demonstrates the interface and sample workflow.
Note: The deployed GitHub Pages website is a static demo and is not connected to the local AI backend. Local AI inference runs separately through the application's local setup.
Code
GitHub repository: https://github.com/utkarshydv27/DukaanFlow-AI
The repository contains the project source code and setup information.
How I Built It
DukaanFlow AI explores a local-first AI architecture using:
- Ollama to run a language model locally.
- Qwen2.5 1.5B for local natural-language interpretation.
- Rule-based fallback handling to support more predictable behavior.
- Seller confirmation before important order changes are applied.
- GitHub Pages to deploy the public demo website.
The idea is to combine the flexibility of an open-weight language model with explicit application rules. Rather than blindly trusting every AI response, the application should validate proposed actions and let the seller confirm important changes.
Why Does Open Innovation Matter?
Open innovation makes it possible to experiment with AI tools that developers can inspect, adapt, and run locally.
For small businesses, a local-first approach can reduce dependence on paid, per-request cloud AI APIs. It can also reduce the need to send customer messages to an external AI provider when processing is genuinely kept local.
Local inference still requires suitable hardware, electricity, and maintenance, and privacy depends on the complete implementation.
Most importantly, open-source tools give developers the freedom to customize AI for specific real-world problems. In this project, that means exploring a solution for small food sellers instead of building another generic chatbot.
My Agent Session
I do not have a public agent-session recording to share for this submission, so I am leaving this optional section out.
Prize Categories
This project is submitted for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle.
I have not claimed any additional partner prize category.
I'd love feedback from developers and small-business owners on how DukaanFlow AI could become more useful in real-world order management.
Live demo: https://utkarshydv27.github.io/DukaanFlow-AI/
Source code: https://github.com/utkarshydv27/DukaanFlow-AI
Top comments (0)