I'm a software engineer at work. Learned by building and experimenting on the internet. Most days go into TypeScript, backend architecture. I share updates and thoughts on X / Twitter.
I build real-time systems, AI apps, and backend infrastructure. You can explore the code on GitHub.
Outside of coding, I like to read books, listen to music, and overthink ideas.
Proof of work
TradeX — Trading Platform
Built a distributed trading system handling live market data, order execution, and real-time account state synchronization.
Built a distributed trading system handling live market data, order execution, and real-time account state synchronization.
- •WebSocket-based streaming for low-latency price updates
- •Event-driven architecture using Redis Pub/Sub for ticker distribution
- •Order execution pipeline updating positions and wallet state atomically
- •Separate services (HTTP, WS, Rust) for scalability and isolation
Pencil — Whiteboard collaboration tool
Built a multi-user whiteboard system with real-time synchronization of drawing actions across clients.
Built a multi-user whiteboard system with real-time synchronization of drawing actions across clients.
- •WebSocket-based event system for broadcasting drawing operations (shapes, strokes, text) in real-time
- •Canvas rendering engine handling pan/zoom transformations and smooth drawing interactions
- •Room-based architecture for isolated multi-user sessions
- •Live cursor tracking and user presence synchronization
Retriv — Website RAG Agent
Built a system for generating domain-specific AI agents using website data, enabling contextual Q&A through retrieval-augmented generation (RAG).
Built a system for generating domain-specific AI agents using website data, enabling contextual Q&A through retrieval-augmented generation (RAG).
- •Pipeline for ingesting and chunking website content, storing embeddings in Qdrant vector database
- •Semantic search layer to retrieve relevant context for user queries
- •LLM orchestration using LangChain for response generation with contextual grounding
- •Embeddable widget with real-time chat interface for seamless website integration
Work Experience
SDE Internat████████
Designed and built an end-to-end speech-to-speech AI calling system for automated user feedback collection
- Engineered real-time audio streaming pipelines over WebSockets, ensuring low-latency bidirectional communication
- Integrated speech-to-text and text-to-speech pipelines, delivering seamless, human-like voice interactions
Skills
Github
Let's Connect
