Screenshot of Limit Order Book Engine (lob-engine)

Limit Order Book Engine (lob-engine)

A C++ engine that reconstructs live crypto-exchange order books in real time and backtests short-horizon trading signals on the rebuilt book.

  • C++
  • Python
  • CMake
  • WebSockets
  • scikit-learn

A from-scratch C++ engine that ingests live Level-2 market data from crypto exchanges (Coinbase, Kraken, Binance.US) and rebuilds the full limit order book in memory. On top of the reconstructed book sit a backtester for order-flow-imbalance signals, a passive market-making simulation with a queue-position fill model, and a short-horizon price-direction model evaluated walk-forward. I profiled and optimized the hot path — cutting per-row parse time about 6x and adding a cache-friendly flat-array book validated by a differential test against the reference — and added a live streaming pipe, Linux/Windows CI, and a zero-dependency browser dashboard that reconstructs any supported book live.

Java HTTP Server From Scratch

A simple HTTP server built in Java using raw sockets

  • Java

Built to understand HTTP and networking at a low level. Uses Java’s ServerSocket to accept raw TCP connections, manually parses incoming request lines and headers, and constructs HTTP responses by hand rather than relying on a framework like Spring.

Maitri

Social-impact Android app connecting families in need with verified donors for essential resources.

  • Java
  • Firebase
  • Android
  • UI/UX

Connected around 25 slum-dwelling families with donors for food, healthcare, education, clothing, and childcare. Streamlined donor matching, partnered with NGOs to validate needs, and expanded outreach to grow beneficiary support.

AI & Robotics in Education (Published Research)

Peer-reviewed literature review on how AI and robotics are reshaping the roles of teachers and students.

  • Artificial Intelligence
  • Machine Learning
  • Ed-Tech
  • Literature Review

Published in the International Journal of Scientific Research in Computer Science, Engineering & Information Technology. Reviewed 48 sources (refined to 25), analyzing the benefits, challenges, and implications of AI-driven education models.