Services & Portfolio
Bridging high-level software architecture with low-level hardware performance, my approach integrates mathematical rigor with high-performance computing to solve complex engineering challenges.
Software Development
I specialize in building robust, scalable applications that require high-density data processing and complex numerical modeling.
- Scalable Microservices Architecture
- Physical/Chemical Software Simulations
- Graphical Software Simulations
- Data Analysis & Applied Machine Learning
- Website Development
- Software Debugging
- Software Testing
- Software Development Best Practices
- Technical Documentation
- Consulting & Advisory Services
Energy Load Forecasting Framework using Python
A comparative analysis framework for testing regression and deep learning algorithms on the Building Data Genome dataset. It predicts hourly electrical consumption using weather data and engineered features like .
Image Processing Suite using C# .NET
Real-time image manipulation demonstrating the bridge between managed C# and unmanaged CUDA C++ kernels via P/Invoke for high-occupancy ARGB matrix convolutions.
Website Development using Typescript + React
A production-grade personal portfolio built with Next.js 16 and React 19, featuring a React Native Web hybrid architecture for cross-platform readiness, scroll-spy navigation via IntersectionObserver, GA4 analytics with cookie consent, and Firebase Hosting deployment.
Backend API Development using Java + SpringBoot
A scalable, high-performance backend solution architected using the Spring Boot framework. This project prioritizes clean design patterns, implementing robust RESTful APIs that facilitate seamless communication between services. The backend incorporates efficient data persistence layers using JPA/Hibernate, strict input validation, and comprehensive security protocols via Spring Security. The result is a modular, maintainable system designed for reliability and ease of integration with any frontend client.
Fullstack Virtual Shop using Typescript + React / Java + SpringBoot
A comprehensive e-commerce platform designed for performance and scalability. This application features secure user authentication, a dynamic product catalog with real-time filtering, and a seamless cart management system. The architecture leverages a responsive React and TypeScript frontend for clients and back-office operations integrated with a robust Java Spring Boot backend, utilizing RESTful APIs to handle complex business logic, secure order processing, and efficient database management.
Hardware & GPU Acceleration
Pushing the boundaries of computational efficiency through custom GPGPU kernels and integrated embedded systems.
- CUDA Kernel Optimization
- Embedded Systems (Arduino / Raspberry Pi)
- IoT Solutions
- Hardware Abstractization Layers
GPU Heat Transfer Simulation using C++ and CUDA
Numerical simulation of 2D transient heat conduction using the Finite Difference Method. Calculates thermal equilibrium using: Where is the total temperature difference between the last 2 computed steps and is the predefined stopping tolerance. It includes a visual representation of the process.
Matrix Convolution Library using CUDA
Standalone CUDA backend for simultaneous 4-channel matrix convolution processing. Extracts 8-bit color channels from 32-bit integers using optimized bitwise shifts to maximize GPU throughput. It exposes the compute API.
Arduino Data Acquisition Module
An intelligent data acquisition system designed to monitor and log environmental parameters in real-time. This module utilizes an Arduino microcontroller interfaced with a suite of high-precision sensors to capture critical telemetry data. The system features on-board storage capabilities for local logging and supports wireless connectivity for remote data transmission, enabling seamless visualization and analysis of hardware-level inputs.