A three-part toolkit for browser session capture, automated crawling, and offline reconstruction of modern web applications.

F1 data analysis & telemetry workstation — lap comparison, throttle/rpm charts, track visualization

Empirical analysis of LEO satellite network handover behavior using real Starlink performance data. Automated collection of 500+ data points revealing 73% connection stability and optimal elevation angles for satellite positioning.

Forza Horizon 6 Deployment Studio is an advanced client optimization pipeline and asset manager designed for the new Japanese festival map.

Production-ready F1 strategy simulator with real-time telemetry analysis, pit optimization, and faster API responses via Redis caching.

A powerful tool for analyzing telemetry data, leveraging machine learning to automatically detect anomalies and unusual patterns. Ideal for monitoring system health, identifying performance issues, and proactive fault detection.

A high-fidelity sequence-to-one Transformer pipeline designed to forecast Formula 1 race outcomes with 95.47% Podium Accuracy. This project represents a complete end-to-end implementation of a deep learning architecture for temporal racing data. I developed a custom 5-layer Transformer Encoder to process 10-lap sliding windows of telemetry context

Python package to help with track data & telemetry data

Advanced Formula 1 data engineering and performance analytics ecosystem. Features MATLAB-based pipelines for tire degradation modeling, fuel-normalization, and automated pre-season testing telemetry processing for the 2026 regulations.

F1 telemetry intelligence system that reconstructs laps meter-by-meter using FastF1 data, compares drivers with millisecond delta alignment, and generates race-engine level insights with interactive visualizations.

Actuator health intelligence for robotics reliability, telemetry analysis, drift detection, and diagnostic reporting.

A Python tool that analyzes the fastest lap between two F1 drivers

Real-time F1 telemetry analysis with a TCN (ONNX) model and React/WebSocket dashboard. Trained on Hamilton & Verstappen lap data via FastF1.

Valorant Telemetry Studio is an advanced input analysis and computer vision pipeline designed for the Riot Client environment. This analytical framework allows competitive players to track real-time pixel rendering, monitor kernel-level input latencies, and analyze crosshair vector telemetry.

Tools and documentation for examining what AI platforms do during your sessions using standard browser developer tools (F12/DevTools). Covers Claude, ChatGPT, Gemini, Grok, DeepSeek, and Perplexity.

A bare-metal TinyML autonomous RC car running an INT8 MobileNetV2 model and sensor fusion on an ESP32-S3, achieving 40–60ms real-time inference latency via FreeRTOS without relying on Linux or the cloud.

Industrial IoT Telemetry Forecasting (IITF) system using a specialized Dual-Trace LSTM architecture and Huber Loss for robust cloud resource demand prediction.

Motorsports telemetry analytics, ML pace forecasting platform and driver clustering using FastF1, Scikit-learn, and Streamlit.

End-to-end telemetry pipeline demo with AI-assisted scenario generation, deterministic synthesis, vector retrieval, and evidence-based summarization.