How kevalparmar75/summit-slayer works
Project type and purpose
• Talent Tuner is a full-stack web application (Flask + optional Streamlit frontend) that ingests user inputs (resume text, code snippets, LinkedIn profile data) and uses Google’s Generative AI (Gemini) API to analyze and return improvement suggestions.
File structure
• Top level:
• app.py – Flask application entry point
• main.py – Streamlit application entry point (alternative frontend)
• codereview.py – Python module containing the logic to call Gemini for code review
• css/, js/, images/ – static assets for the Flask‐served UI
• templates/ – Jinja2 HTML templates (index.html, login.html)
• requirements.txt – lists Python dependencies
README for additional insights
• Key features: Resume Review, Code Review, LinkedIn Profile Review
• Tech stack: Flask (web server), Python modules with logic, Streamlit as an optional quick frontend, external Google Gemini API
• No database is mentioned—state is ephemeral or passed through API calls.
Elements
• Labels:
• “Browser UI” / “Streamlit UI”
• “Flask App (app.py)”
• “codereview.py”, “resumereview.py”, “linkedinreview.py”
• “Google Gemini API”
• Flow of data (User → Flask → Logic → Gemini → Flask → User)
• If using Streamlit: User → Streamlit → Logic → Gemini → Streamlit → User
• Rectangles for components (UI, server, modules)
• Clouds for external services