AI-Powered Proof of Work

Proofly is an AI-powered platform that analyzes GitHub projects and tests developers' technical knowledge, helping them showcase real skills through verifiable evidence, not just resume claims.

Comment

Proofly — AI-Powered Proof of Work

Don't Just Claim It. Prove It.

Inspiration

In today's AI-driven world, building a project has become easier than ever. But proving that you actually understand what you've built? That's a completely different challenge.

GitHub profiles and resumes can showcase impressive projects, but they rarely demonstrate a developer's actual technical understanding or individual contributions.

We wanted to bridge this gap.

What if your projects could speak for your skills? What if your technical knowledge could be demonstrated through the very code you've written?

This question inspired Proofly, an AI-powered platform that aims to transform ordinary GitHub repositories into evidence-backed demonstrations of technical ability.

What it does

Proofly is designed to help developers showcase their skills through project-specific technical assessments.

The platform follows a simple workflow:

GitHub Repository → AI Analysis → Technical Assessment → Proof of Work Report

Key Features

  • GitHub Repository Analysis: Examines repository structure, technologies, documentation, and available contribution history.
  • AI-Generated Technical Questions: Generates questions based on the project's actual implementation rather than generic programming concepts.
  • Intelligent Skill Assessment: Evaluates developers' explanations against available repository context and provides constructive feedback.
  • Proof of Work Reports: Creates shareable reports highlighting project details, assessment results, and supporting technical evidence.
  • Developer Portfolio: Presents completed assessments and project information in a professional format.

Instead of simply asking developers what they know, Proofly aims to let them demonstrate their understanding.

How we built it

Proofly is designed around a modular web architecture that connects GitHub repository data with generative AI.

Technology Stack

  • Next.js: Frontend development and application routing.
  • Tailwind CSS: Responsive and modern user interface.
  • GitHub REST API: Repository metadata, file structure, and contribution information.
  • Google Gemini API: Repository analysis, technical question generation, and assessment feedback.
  • Supabase: Authentication and data storage.
  • Vercel: Web application deployment.

Development Approach

Our development approach focuses on creating a complete user journey rather than disconnected AI features.

First, the application retrieves relevant repository information through the GitHub API. The extracted context is then prepared for AI analysis.

Next, the AI generates technical questions focused on implementation decisions, architecture, and programming concepts demonstrated in the repository.

The developer responds to these questions, and the system evaluates the explanations using repository context.

Finally, the platform organizes the assessment results into a structured, shareable report.

The objective is to make technical skill demonstration accessible, transparent, and practical.

Challenges we ran into

1. Understanding Different Codebases

Every GitHub repository has a different structure, technology stack, and level of documentation.

Designing an analysis system that can extract meaningful technical context without processing unnecessary files is a significant challenge.

2. Generating Relevant Questions

Generic programming questions wouldn't make Proofly different from existing interview preparation platforms.

The challenge is ensuring that generated questions relate directly to a project's implementation and require genuine technical understanding.

3. Evaluating Answers Fairly

Technical concepts can be explained in multiple valid ways.

An effective assessment system needs to evaluate conceptual correctness rather than rely on exact wording or keyword matching.

4. Balancing Trust and AI Limitations

AI cannot conclusively establish who authored a repository simply by analyzing code or questioning a developer.

We therefore focused our design on evidence-based technical assessments rather than claiming absolute proof of authorship.

5. Keeping the Experience Simple

Repository analysis and AI evaluation involve multiple technical processes. Presenting them through a clean, intuitive interface is essential for making the platform useful to students and recruiters.

Accomplishments that we're proud of

We're proud of developing an idea that approaches technical skill verification from a different perspective.

Rather than creating another AI resume builder or generic interview chatbot, Proofly connects assessments directly to developers' own projects.

Our key design accomplishments include:

  • Defining an end-to-end workflow for repository analysis and technical assessment.
  • Designing personalized questions around real project implementations.
  • Developing an approach to evidence-backed skill reporting.
  • Prioritizing transparency and responsible AI evaluation.
  • Creating a product concept with applications in recruitment, education, and developer portfolios.

Our biggest accomplishment is shaping Proofly around a simple belief:

A developer's demonstrated understanding should matter more than the number of technologies listed on their resume.

What we learned

Working on Proofly has helped us explore how generative AI can support more meaningful technical evaluation.

We learned that understanding a software project involves more than identifying programming languages or summarizing code. Architectural decisions, trade-offs, and implementation reasoning are equally important.

We also recognized the importance of designing AI systems that communicate their limitations.

An automated assessment should provide useful evidence and feedback without presenting uncertain conclusions as facts.

Most importantly, this project reinforced the value of solving a genuine problem rather than adding AI features simply for the sake of using AI.

What's next for AI-Powered Proof of Work

Proofly has the potential to evolve into a broader technical portfolio and skill assessment ecosystem.

Our planned improvements include:

  • Live Technical Interviews: Interactive AI follow-up questions based on repository implementations.
  • GitHub Contribution Insights: Deeper analysis of commit history and collaborative development.
  • Recruiter Dashboard: A dedicated interface for reviewing project evidence and technical assessments.
  • AI Portfolio Recommendations: Personalized suggestions for improving documentation and project presentation.
  • Multi-Platform Support: Integration with GitLab and other code hosting services.
  • Team Contribution Analysis: Additional tools for understanding collaborative project work.

Our long-term vision is to make Proofly a place where technical skills are demonstrated through meaningful evidence, not just claimed through credentials.

Proofly — Because real skills deserve real recognition.

Built With

Share this project:

Updates

Submission history