This is a submission for Weekend Challenge: Passion Edition
What I Built
Most career platforms evaluate what you've done.
I wanted to build something that explores why you do it.
PassionDNA AI is an AI-powered self-discovery platform that combines:
GitHub portfolio analysis
Personalized AI interview
LLM reasoning
Evidence-backed passion analysis
Instead of asking users to complete another personality quiz, PassionDNA compares what they say they're passionate about with what they've actually built.
The final result is an explainable PassionDNA Report containing:
Passion Confidence Score
Passion Categories
Hidden Strengths
Growth Opportunities
GitHub Evidence
AI Summary
Personalized Challenge Question
The goal is simple:
Help developers discover whether their daily work truly reflects what they love building.
Demo
🌐 Live App
https://passiondna-ai-havgehgxkzqrjyeqg4g2bs.streamlit.app/
💻 GitHub Repository
https://github.com/uroojbuilds/passiondna-ai.git
Screenshots
Landing Page
The application introduces the PassionDNA concept and explains how GitHub activity, AI interviews, and reasoning combine into one report.
User Information
Users provide:
GitHub username
About Me
This information personalizes the AI interview.
AI Passion Interview
Instead of generic personality questions, Gemini generates five thoughtful interview questions designed to reveal genuine motivation.
Examples include:
Which project made you lose track of time?
What problem would you solve even if nobody paid you?
What impact do you want your work to have?
Personalized Questions
Every interview focuses on real experiences rather than multiple-choice answers, making the final report much more authentic.
PassionDNA Report
The report includes:
Passion Confidence
Passion Categories
Visual Analytics
Evidence-Based Insights
Every recommendation is backed by GitHub evidence instead of vague personality descriptions.
Users receive:
Hidden Strengths
Growth Opportunities
Evidence Behind Every Insight
AI Summary
Finally, Gemini creates an explainable summary of the user's engineering profile and ends with a personalized challenge designed to encourage future growth.
How I Built It
Tech Stack
Python
Streamlit
Google Gemini API
GitHub REST API
HTML/CSS
Custom Components
Workflow
GitHub Username
│
▼
GitHub API Analysis
│
▼
About Me Analysis
│
▼
Gemini Generates
5 Personalized Questions
│
▼
Interview Answers
│
▼
Gemini Reasoning
│
▼
PassionDNA Report
Key Features
🧠 AI Interview
Google Gemini dynamically generates thoughtful interview questions based on the user's GitHub profile and personal description.
💻 GitHub Portfolio Analysis
The app analyzes public repositories to identify:
Languages
Repository Topics
Technologies
Project Patterns
🔍 Explainable AI
Rather than producing a random score, every conclusion is supported by GitHub evidence and interview responses.
📊 PassionDNA Dashboard
The final report includes:
Passion Confidence
Passion Categories
Verification Summary
Hidden Strengths
Growth Opportunities
AI Summary
🛡 Graceful Fallbacks
If Gemini is unavailable, the application automatically switches to predefined interview questions and rule-based report generation, ensuring the demo continues to work.
Biggest Challenge
One of the hardest parts was ensuring the application never failed during the user journey.
Because AI APIs can occasionally fail due to limits or network issues, I implemented fallback logic so the interview and report generation still work even without a successful Gemini response.
This made the application much more reliable for demonstrations.
What I Learned
This project taught me that explainability matters just as much as intelligence.
Users trust AI more when it can clearly show why it reached a conclusion instead of simply providing a score.
Combining structured GitHub evidence with conversational AI created a much more transparent experience than a traditional personality quiz.
Future Improvements
If I continue developing PassionDNA AI, I would like to add:
LinkedIn integration
Resume analysis
AI voice interview
Personality evolution over time
Team PassionDNA comparison
Career roadmap generation
Multi-agent reasoning
Deployment recommendations
Prize Categories
✅ Best Use of Google AI
Google Gemini powers:
Personalized interview generation
AI reasoning
Passion report generation
Final summaries
Personalized challenge questions
Building PassionDNA AI was an exciting opportunity to explore how AI can move beyond answering questions and instead help people better understand themselves. I hope this project demonstrates how LLMs can be used not just for productivity, but for meaningful self-discovery.
Thank you for reading! 🚀
Top comments (0)