FixFlow: Small Spaces. Clear Requests.

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

🛠️ FixFlow: Small Spaces. Clear Requests.

Imagine managing a hostel or PG where residents report broken taps, faulty fans, leaking pipes, and electrical problems through scattered messages.

One person sends a text. Another gives an incomplete description. Someone else reports the same problem again. The manager has to make sense of all of it before a repair can even begin.

FixFlow is built to make that process less chaotic.

It's an AI-powered maintenance request management application designed for hostels, PGs, and other shared living spaces. It helps transform everyday complaints into structured requests that managers can review, organise, and track.

Our goal is simple: make maintenance easier to report, easier to understand, and easier to manage.

🏠 What I Built

FixFlow brings the resident's complaint and the manager's workflow into one application.

✍️ 1. Turn everyday complaints into structured requests

Residents can describe a problem in their own words instead of filling out complicated forms.

For example:

“The bathroom tap keeps dripping even when it is closed.”

FixFlow is designed to help turn descriptions like this into structured maintenance information. Residents can provide building, floor, and room or common-area details, while unknown location fields can be left blank.

The request can then be reviewed before it is sent to the manager.

🧠 2. AI-assisted complaint processing

Maintenance descriptions aren't always complete or consistently written. AI can help extract useful information, identify missing details, and recommend how a request should be handled.

FixFlow includes a Gemma-powered extraction option through Ollama, alongside a demo mode that displays sample results.

The intention is not to let AI make every decision independently. Instead, it helps organise the information so that a human can review the request before acting on it.

📋 3. A dedicated manager dashboard

The manager dashboard provides a separate entry point for reviewing resident requests, assignments, and activity history.

The project is designed around a practical workflow: collect the complaint, structure the information, review it, and manage the maintenance process.

🎯 Who is FixFlow for?

• Hostel and PG owners managing multiple rooms.
• Small property managers who need a clearer maintenance workflow.
• Staff responsible for handling recurring repair requests.
• Shared accommodation operators who want fewer scattered complaints and better visibility.

A tool that saves management time and helps prevent requests from getting overlooked could offer real value to these users.

🖥️ Demo

The screenshots below show the two main views of the application.

1. New Maintenance Request

The resident-facing interface lets someone describe an issue, provide available location details, select the extraction mode, and review the request before submission.

2. Manager Dashboard

The manager-facing interface provides a dedicated sign-in screen for accessing resident requests, assignments, and activity history.

These views illustrate the intended experience of making maintenance requests more structured and manageable.

💻 Code

🔗 Project repository: https://github.com/abirami-prabhakar/flowfix

The repository contains the project's implementation and setup instructions.

⚙️ How I Built It

FixFlow combines a lightweight Python application with open-source AI tooling.

The technology stack:

• Python — application logic.
• Streamlit — interactive application interface.
• Gemma via Ollama — open-weight language model for AI-assisted complaint processing.
• Pydantic — structured data validation.
• SQLite — data storage.
• Sentry — optional error monitoring.

The application is designed to accept a resident's description, process the information into a structured format, and let the user review the result.

Pydantic supports structured validation, while SQLite provides a lightweight database option for application data. The interface separates the new-request workflow from the manager dashboard.

The project also offers a demo mode with sample results. This makes it possible to explore the interface without depending on a live model response for every demonstration.

🌱 Why Does Open Innovation Matter?

For me, the interesting part of FixFlow isn't simply adding AI to a form. It's exploring how open-source AI can make a practical, focused tool more accessible to smaller organisations.

A hostel or PG doesn't necessarily need a massive AI platform. It needs a reliable way to organise everyday maintenance problems.

Open-weight models and open-source tooling give developers more freedom to experiment with that kind of focused solution.

💰 1. More control over operating costs

A local model can reduce dependence on paid, per-request AI APIs. For a small property-management tool, controlling recurring costs matters.

Local inference still requires suitable hardware and electricity, so it isn't automatically free. But it gives developers another way to manage the cost of running AI.

🔐 2. More control over data

Maintenance requests can contain information about residents, rooms, and shared spaces.

Using a locally hosted model can reduce the need to send complaint text to an external AI provider. That can be valuable when privacy matters, provided the rest of the application is configured appropriately.

Local inference alone does not guarantee that every part of the application stays private.

🔧 3. Freedom to experiment

Open-weight models make it easier to experiment with model configurations and explore alternatives for a specific domain.

For FixFlow, that means we can investigate how well a model understands maintenance-related language, how it structures complaints, and where human review is still necessary.

🌍 4. Building for smaller organisations

Open-source AI makes experimentation more accessible to developers who want to solve practical problems without making a proprietary AI service the foundation of everything.

That's the open-innovation idea behind FixFlow: use adaptable AI to solve a small, concrete problem, while keeping humans involved in decisions that affect real people.

🤝 Team Credits

FixFlow was developed collaboratively with my friend @abirami_prabhakar.

We explored how open-source AI could help transform everyday maintenance complaints into structured, actionable requests. This project reflects our collaborative effort to build a practical application around an open-weight model.

🔗 Original GitHub repository: https://github.com/abirami-prabhakar/flowfix

🚀 What's the real opportunity?

For a hostel or PG manager, the value isn't AI for its own sake.

It's fewer scattered complaints, clearer information, and a more organised way to handle maintenance.

FixFlow explores how a small, focused AI application could help make that happen.

If you managed a hostel or PG, would you use a tool like FixFlow to handle maintenance requests? What would it need to do for you to consider paying for it?

I'd love to hear your thoughts!