cloud-learning-journey

From fundamentals to deployment.

This repository documents my projects and progress as I follow a strategic roadmap to land a top-tier tech role.

Completed the Introduction to IAM lab. Learned about IAM roles, policies, and service accounts.

First Production Deployment on Cloud Run

🎉 MAJOR MILESTONE ACHIEVED

Live Application URL: https://my-first-cloud-app-640781293504.us-central1.run.app

What I Accomplished

1. Set up secure GCP environment with budget controls
2. Created Docker container from scratch
3. Debugged real deployment issue (port configuration)
4. Successfully deployed to Google Cloud Run
5. Application is now live and publicly accessible

🛠 Technical Stack Used

• HTML/CSS/JavaScript
• Docker Containerization
• Google Cloud Run
• Google Cloud Build
• Nginx Web Server

Problem Solved

Initial Deployment Failed: Container port misconfiguration Root Cause: Nginx listening on port 80, Cloud Run requires port 8080 Solution: Modified nginx configuration to listen on port 8080

📚 Key Learnings

• Cloud Run requires containers to listen on port 8080
• Dockerfile configuration is critical for deployment success
• Debugging deployment issues is a valuable skill
• Professional deployment workflow: local → container → cloud

🎯 Mission Accomplished

Live AI Sentiment Analyzer with Database Integration

URL: https://ai-sentiment-analyzer-640781293504.us-central1.run.app

✅ What Works

1. Gemini AI Integration - Real sentiment analysis via models/gemini-2.0-flash
2. Database Persistence - SQLite with automatic saving to /tmp
3. Full REST API - 4 endpoints with proper JSON responses
4. Production Deployment - Google Cloud Run with auto-scaling
5. Frontend Interface - Browser-accessible UI

🔧 Technical Stack

• Backend: FastAPI 0.104.1
• AI Model: Gemini 2.0 Flash
• Database: SQLite (persistent in Cloud Run /tmp)
• Deployment: Google Cloud Run (serverless)
• Memory: 512Mi
• Region: us-central1

🧪 Test Results

Input: "Gemini model fixed! Database integration complete!" Output:

• Sentiment: positive (95% confidence)
• Key Phrases: ["Gemini model fixed", "Database integration complete"]
• Database Save: ✅ Successful
• Response Time: < 3 seconds

🚀 Features Live

• POST /analyze - AI sentiment analysis + database save
• GET /history - Retrieve chronological analysis history
• GET /stats - Analytics (sentiment distribution, averages)
• GET /health - Service health + database connection check
• GET / - Frontend interface with sentiment input

This project demonstrates:

• Full-stack development from idea to production
• Cloud deployment expertise (Google Cloud Run)
• AI integration skills (Gemini API with proper model handling)
• Database design (SQLite with schema and queries)
• Production debugging (solving real deployment issues)
• API design (RESTful endpoints with proper responses)

🚀 Live Production Application

2. ML Model Monitoring System

URL: https://ml-model-monitor-640781293504.us-central1.run.app Features: Production ML monitoring, drift detection, alerting system, Streamlit dashboard

📊 What These Projects Demonstrate

Technical Skills:

• FastAPI for production backend APIs
• MLOps with model monitoring and drift detection
• Cloud Deployment on Google Cloud Run
• Database Design with SQLAlchemy
• Authentication & Security with API keys
• Data Visualization with Streamlit and Plotly

🔧 Tech Stack Mastery

• Backend: FastAPI, SQLAlchemy, Pydantic
• ML/AI: Gemini API, scikit-learn, statistical testing
• Database: SQLite (production-ready patterns)
• Cloud: Google Cloud Run, Docker, gcloud CLI
• Monitoring: Custom metrics, health checks, alerting
• Visualization: Streamlit, Plotly