I'm currently 26 years old and I'm a hands-on software engineer with industry experience developing desktop applications. I have comprehensive knowledge about Agile Methodologies and programming architectures. I once was a Desktop application lover, but boy-oh-boy the web development environment got me good! But I still have thorough knowledge of .NET and .NET Core.
An event-driven, low-latency FX order scheduling system that enables clients to schedule automatic execution of FX spot orders when the market (treasury) rate reaches a client-visible target.
The architecture is designed to:
- Match treasury ticks to scheduled orders in real-time
- Claim and execute orders with exactly-once semantics
- Scale to millions of scheduled orders while maintaining bounded latency
- Provide event-driven processing for high throughput
- Infrastructure: AWS with Terraform (Lambda, DynamoDB, SQS, SNS, ElastiCache Redis, VPC)
- Architecture: Event-driven serverless design with low-latency Redis caching
- Focus: Distributed systems, exactly-once processing, high-scale event matching
A self-hosted, open source feature flag service built with .NET 9, designed for teams who want full control over their release toggles without relying on SaaS vendors. Deployable on AWS via Terraform and Kubernetes Helm charts.
Key features include:
- Remote evaluation API for checking feature enablement
- Admin API for managing organizations, projects, environments, and feature rules
- SignalR WebSocket hub for real-time toggle change notifications
- PostgreSQL as source of truth with Redis for low-latency caching
- Back-end: .NET 9, ASP.NET Core, SignalR for real-time communication
- Infrastructure: AWS (EKS, RDS PostgreSQL, ElastiCache Redis), deployed via Terraform and Helm
- Architecture: Modular, event-oriented, API-first design
- Focus: Self-hosting, enterprise features, distributed caching
Implementation of the CLEAN architecture in a production-like C# solution integrating with a weather HTTP API. Includes Prometheus and Grafana for monitoring and observability. Built with ASP.NET Core, Serilog, and Prometheus.NET.
A simple chess bot using Lichess and Stockfish engine, built with Python.
A modern support ticket handling system built with Laravel (back-end) and React with Tailwind CSS (front-end). Designed to offer a faster, more intuitive experience than traditional internal ticketing systems.
An AI fruit recognition system built with TensorFlow and Python using the Fruits-360 Dataset. Explores LSTM and GRU layers for image classification.