Projects
Here are some of the projects I've worked on.
Local-Vibes Event Management Platform
Built a full-stack local event discovery and management platform using Go with Chi router, Vue.js 3, and SQLite. Implemented secure authentication, fuzzy search with Supabase backend, real-time event CRUD operations, and personalized dashboards enabling 20+ beta testers to create, discover, and join community events with custom rules and capacity management.
Chat Application
Designed a scalable chat platform for an Advanced OS course project inspired by Discord using Go, WebSockets, and implementing consistent hashing and rate limiting across 8 nodes, which enabled real-time communication for 30+ testers and demonstrated potential to scale to 2,500+ users. deployed on AWS with Docker and CI/CD, supporting clients with low as 10ms latency and 40% faster deployment.
Tweets Sentiment Analysis
Developed a high-performance sentiment analysis engine for an Introduction to Big Data course by leveraging Llama-2, LangChain, Apache Spark, and AWS to process 10,000+ tweets/hour with 92.2% accuracy, delivering real-time insights via a responsive dashboard and reducing processing time by 30%.
Realtime Multiplayer Tic Tac Toe
Designed a Realtime multiplayer Tic-Tac-Toe game that can be played from multiple devices.
MangaFlow Reader
Built a full-stack manga reading platform using Go with Chi router, featuring MangaDex API integration, concurrent image processing with goroutines, in-memory caching, and a responsive Tailwind CSS interface supporting search, popular/recent manga listings, chapter navigation, and Docker deployment.
AI Recipe Generator
Developed a serverless AI-powered recipe generator using React, TypeScript, AWS Amplify, and Claude 3 Sonnet via AWS Bedrock. Implemented user authentication with AWS Cognito, GraphQL API for recipe generation, and custom resolvers for AI integration, delivering personalized recipe suggestions from user-provided ingredients.
LLM CLI Chat Bot
Built a versatile command-line AI chat interface in Go using Cobra CLI framework and LangChain, supporting multiple LLM providers including OpenAI GPT, Google Gemini, and local Ollama models with streaming responses, conversation history management, and graceful interrupt handling.