This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle.
What I Built
Some projects start with a brilliant business idea. Others start with a problem you keep thinking about.
ProjectIQ started with me thinking about how much work goes into planning a project and evaluating a tender.
My father was part of what made me think about real-world work and the decisions people have to make. It made me curious about something: what if evaluating a project didn't have to mean spending hours going through documents, checking requirements, and trying not to miss something important?
I didn't have a perfect answer. I had Python, a lot to learn, and an idea that seemed worth exploring.
That's how ProjectIQ began.
ProjectIQ is an AI-powered project and tender analysis platform I'm building to help contractors and project-based businesses evaluate documents, identify missing information, spot contradictions, and understand whether they have the resources to take on a project.
Think about a tender with dozens of pages of requirements, equipment specifications, and other details. Missing one important requirement could create problems later. ProjectIQ aims to help make those details easier to examine.
The long-term vision is to bring document analysis, requirement matching, equipment and workforce assessment, financial analysis, and risk evaluation into one workspace.
Is the entire platform finished? Not even close.
The backend foundation is in place, and I've implemented and tested individual components for requirement validation, quotation verification, contradiction detection, and equipment analysis. Local AI integration through Ollama is also part of the project.
Other pieces, including the complete document-to-analysis workflow, financial analysis, frontend integration, and reporting, are still being developed.
I'm sharing ProjectIQ at this stage because I want to build it openly, learn from other developers, and get feedback before assuming I know exactly what users need.
Who would pay for this?
Potential users include small and medium-sized contractors, consultants, and project-based businesses that spend significant time evaluating tender documents.
If ProjectIQ can eventually reduce repetitive checking and help teams identify important gaps earlier, it could become a useful business tool. Possible future models include self-hosted deployments, paid setup and customization, or team-oriented features.
That's the opportunity I'm exploring—not a claim that I've already found paying customers.
Demo
ProjectIQ is still under development, so I don't have a complete deployed demo to share yet.
For now, you can explore the code and follow the progress here:
đź”— GitHub repository: https://github.com/asvibes/ProjectIQ
I'll add a demonstration when the end-to-end workflow is ready.
Code
ProjectIQ is being developed as an open-source project with a modular structure.
The repository includes the backend, frontend, analysis and verification services, automated tests, and supporting development resources.
I'm trying to build more than a collection of AI prompts. I want the system to have a foundation for validating information, checking evidence, and making its findings easier to examine.
There is still plenty to improve, and that's one reason I want the project to be open to contributions.
How I Built It
My current stack includes:
- Python + FastAPI: The backend foundation.
- Ollama: Running AI models locally.
- SQLAlchemy: Database interaction.
- Pytest: Automated testing.
- React + TypeScript: Frontend development.
- Tailwind CSS: Interface styling.
One of my main goals is to explore what can be done with locally executed, open-weight AI models rather than making a paid, hosted AI API a requirement for every analysis.
The verification and analysis modules are being developed separately so they can be tested before being connected into the complete workflow.
I've also learned that writing a module and integrating it into a working product are two very different challenges.
A test can pass, a function can work, and the application can still have ten other things waiting to break.
Software engineering is humbling like that.
Why Does Open Innovation Matter?
For a project like this, open innovation matters for more than just making the code publicly available.
Tender documents can contain commercially sensitive information. Local AI execution offers a way to explore workflows in which documents stay on a user's own machine or infrastructure, depending on how the system is configured.
Open-source development also gives other developers the opportunity to inspect the implementation, improve the analysis logic, contribute new capabilities, and adapt the platform to their own needs.
I can learn from existing tools instead of reinventing everything, while other people can learn from or build upon what I create.
I don't think open source means every problem is automatically solved. It means more people have the opportunity to participate in solving it.
And as a student still learning how to build useful software, that opportunity matters to me.
What's Next?
My next steps are to connect the existing modules into a more complete workflow, improve document processing, finish the frontend integration, and develop financial analysis and reporting capabilities.
I also want to get feedback from people who actually deal with project planning and tender evaluation. Their experience should help determine which features deserve attention first.
For now, ProjectIQ is a work in progress, not a finished product. I'm documenting that openly because I want the project to grow around real needs rather than an impressive feature list.
Maybe one day, someone will use ProjectIQ and say it saved them hours of work.
For now, I'm the developer trying to make that possible—one module, one test, and occasionally one frustrating error at a time.
And yes, Dad, if this works out, I'll tell people you helped inspire the idea.
I'll keep the Python exceptions to myself.
Top comments (0)