Back to Home

Selected Work

A curated showcase of my most complex, highly-scalable engineering projects.

Next.jsPythonFastAPIWebSockets

Live Market Bot

An advanced cryptocurrency algorithmic trading engine with real-time analytics dashboard.

The Challenge

The goal was to build a highly optimized, extremely fast algorithmic trading bot capable of making split-second decisions using advanced mathematical models (like Bayesian Optimization). The system needed to process massive amounts of candlestick data via WebSockets and present it securely on a beautiful, ultra-responsive dashboard.

The Solution

I architected a dual-stack solution. The backend is a robust Python FastAPI engine deployed via Docker, handling secure Apex API connections, AI Insights generation, and real-time ledger tracking.

The frontend is a blazing-fast Next.js application that leverages custom charting libraries and fluid animations to give users a premium experience with instantaneous theming and live data updates.

Key Features

Real-time AI Insights

Live feed of algorithmic decisions and confidence scores.

Secure Key Management

Enterprise-grade environment variable injection.

Bayesian Tuning

Automated hyperparameter optimization via Optuna.

Dynamic Charting

Responsive SVGs with auto-adjusting themes for Android/iOS.

Project Details

RoleLead Full Stack Developer
StatusActive Pipeline
View Live Dashboard
ReactLeafletFastAPIOpenSky APIDocker

RTA Flight Tracker

A real-time global aviation tracking platform using live ADS-B telemetry streams.

The Challenge

Tracking thousands of high-speed aircraft across the globe requires ingesting massive amounts of state vectors per second. The client needed a robust mapping interface that could render thousands of moving points dynamically without degrading browser performance, paired with a reliable backend that manages strict OpenSky API rate limits via OIDC authentication.

The Solution

I engineered an asynchronous FastAPI backend utilizing `httpx` to continuously poll and process OpenSky network data. The server normalizes the telemetry and broadcasts it via high-throughput WebSockets.

The frontend utilizes React coupled with Leaflet to deliver 60FPS marker interpolations. To prevent UI lockups, data streaming is debounced and mapped efficiently, allowing users to select individual planes, view live altitude parameters, and track flight paths directly over a beautiful monochromatic map skin.

Key Features

Live Interpolation

Smooth visual movement of flight markers mapped using WebSockets.

Global OIDC Auth

Automated token rotation for uninterrupted data ingestion.

Asynchronous I/O

FastAPI handles thousands of concurrent data pipelines flawlessly.

Responsive Map UI

A beautifully styled dark-mode mapping interface.

Project Details

RoleArchitecture & Full Stack
StatusLive Production
Launch Flight Tracker