🚀 Tech Overview – AI-Powered Online Resume#
🎯 Live Project: View Online Resume
This isn’t your average online resume. It’s a serverless, microservices-based project showcasing real-world backend, AI, and cloud engineering skills—all built to run at zero cost (thanks to free tiers and efficient design).
If you’re into geeky stuff, you’re in for a treat!
This isn’t just another run-of-the-mill online resume—nope, we’re taking things to the next level! Instead of a boring static site, this project is a fully-fledged, AI-powered backend system showcasing real engineering skills.We’re embracing a serverless microservices architecture, running both the Golang (Echo) and Python (FastAPI + Gemini) services on AWS Lambda. The Go service handles structured resume data, while the Python service provides AI-powered chat responses.For AI-powered insights, the Python Lambda service integrates with Gemini, using semantic search with Pinecone and data stored in MongoDB. All secrets are securely managed using AWS Systems Manager Parameter Store.As for storage? MongoDB Atlas (free tier) keeps things scalable and cost-effective. Embeddings are stored in Pinecone for lightning-fast semantic search.
🧭 Architecture Flow#
⚙️ Tech Stack Overview#
🔁 Architecture Flow Summary#
1.
Built visually with Firebase Studio
Pushed to GitHub, deployed to Vercel
Interacts with backend via AWS API Gateway
2.
Resume Data Service (Go + Echo)
GenAI Chat Service (Python + FastAPI)
3.
Resume Data Service (Go + Echo)Fetches data from MongoDB Atlas
Manages resume data (personal informations, experiences, etc.)
Loads secrets securely from AWS Systems Manager Parameter Store
4.
Gen AI Chat Service (Python + FastAPI)Processes natural language queries
Uses Gemini API for LLM responses
Embeds resume content semantically
Stores embeddings in Pinecone Vector DB
MongoDB used for storing structured resume data and syncing embeddings
5.
Semantic Search & AI ResponseFastAPI service retrieves relevant context from Pinecone
Generates contextual responses using Gemini
Enables real-time AI chat experience for visitors
📦 Technologies Used#
| Layer | Tech Stack |
|---|
| Frontend | Firebase Studio → GitHub → Vercel |
| API Gateway | AWS API Gateway |
| Serverless | AWS Lambda |
| Resume Backend | Golang + Echo Framework |
| Secrets Manager | AWS Systems Manager Parameter Store |
| AI Service | Python + FastAPI + Gemini |
| Embeddings DB | Pinecone (Vector DB) |
| Resume DB | MongoDB Atlas |
🛡️ Security#
All secrets (API keys, DB credentials) are loaded via AWS Systems Manager Parameter Store
No hard-coded secrets anywhere in the codebase
MongoDB is protected with IP allowlists and access keys
🧠 AI-Powered Interaction#
Resume content is embedded and stored in Pinecone for fast vector search
On-the-fly querying and response generation using Gemini via FastAPI
All processing handled by a separate, isolated Python service (microservice model)
💰 Zero Cost Deployment#
All services used are on free tiers:
MongoDB Atlas: shared cluster
Vercel: personal projects
Gemini: using free API quota
⚠️ Note: Real-world apps at scale may incur costs.
Modified at 2025-05-05 09:02:27