Software Engineer · .NET & Full-Stack Development 📍 United Kingdom
LinkedIn • LeetCode • HackerRank • Codeforces
I'm a backend-focused engineer with full-stack experience, working mainly with .NET, ASP.NET Core, PostgreSQL, Redis, React, and TypeScript. My interest sits at the intersection of software architecture, distributed systems, and making ideas into software that actually works under real conditions — not just in a tutorial.
I'd rather build something that solves a real problem, badly at first, and then fix it properly, than build another to-do app. That's the thread connecting everything below: an observability platform, a video streaming backend, a task manager with real test coverage, a Docker Compose visual editor, an educational HFT simulator, and an algorithm visualizer covering 138+ algorithms.
Outside of shipped projects, I've solved almost 3,200 algorithmic problems across LeetCode, HackerRank, and Codeforces — not to grind interviews, but because working through hard problems regularly is, in my experience, the fastest way to get better at engineering judgment.
- SaaS-oriented backend systems
- Open-source engineering projects
- Deeper skills in distributed systems and advanced PostgreSQL
- Languages: C#, C++, TypeScript, JavaScript, Python
- Backend: ASP.NET Core, .NET, EF Core, Dapper, Node.js, Express.js, SignalR, REST APIs, JWT
- Databases: PostgreSQL, MySQL, Redis
- Frontend: React, Next.js, TypeScript, TailwindCSS, Vite
- Architecture & Tools: Clean Architecture, SOLID, DDD, System Design, Docker, Git, GitHub
A real-time observability platform for log ingestion, metrics aggregation, anomaly detection, and AI-assisted remediation.
Highlights
- Redis-backed processing pipelines with SHA256 deduplication
- Lock-free, thread-safe metrics aggregation using concurrent collections
- Live SignalR dashboards and AI-driven diagnostics with automated patch generation
- Dockerized infrastructure end to end
Stack: .NET 10 · ASP.NET Core · SignalR · Redis · PostgreSQL · Dapper · EF Core · React · TypeScript · Docker · Python Repo: github.com/ezpectus/AI-Log-Observability-MVP
A full-stack video-sharing platform exploring real backend challenges — streaming, transactional consistency, and scalable interaction tracking. I led technical implementation across a six-person team.
Highlights
- HTML5 video streaming via HTTP 206 Partial Content, with local upload and YouTube embed support
- JWT auth with an extensible OAuth-ready architecture
- Atomic likes and subscriptions through database transactions
- View counting optimized with buffered writes to reduce DB load
Stack: Node.js · Express.js · TypeScript · Prisma ORM · PostgreSQL · Next.js · React Repo: github.com/ezpectus/VideoHub
Started as a sandbox for testing backend architecture and design patterns; grew into a complete full-stack app with auth, Kanban, and analytics.
Highlights
- Clean Architecture + Domain-Driven Design throughout
- JWT auth, role-based authorization, structured logging, rate limiting
- 75+ unit tests using xUnit and Moq
- React + TypeScript frontend with Kanban board and analytics dashboard
Stack: ASP.NET Core · .NET · PostgreSQL · EF Core · React · TypeScript Repo: github.com/ezpectus/TaskManager_App
A visual editor for Docker Compose files that actually respects your YAML. Edit the graph, get valid YAML back — comments and formatting intact.
Highlights
- Bidirectional YAML ↔ graph editing
- AST-preserving edits — no structural or formatting loss on round-trip
- Multi-file compose support
Stack: React · TypeScript · React Flow Repo: github.com/ezpectus/docker-compose-visualizer · Live: docker-viz.netlify.app
An educational high-frequency trading simulator built to learn — and demonstrate — real quantitative finance and low-latency engineering techniques. Zero real money, zero real risk, entirely educational.
Highlights
- C++20 signal engine with lock-free SPSC ring buffers, cache-line alignment, branchless programming
- Signal Engine V3 — HMM regime detection with online Baum-Welch adaptation and Viterbi decoding
- 75+ quantitative models (GARCH, Kalman filters, Almgren-Chriss optimal execution, and more)
- ONNX ML inference in C++, a Rust order executor, CUDA-accelerated batch computation
- Shared-memory IPC between C++ and Python, sub-millisecond regime switching
- 191+ live React dashboard panels, full CI/CD across C++/Python/Rust/JS, Kubernetes/Helm deployment
Stack: C++20 · Python · Rust · CUDA · React · Docker · Kubernetes Repo: github.com/ezpectus/HFT-trading-simulator
A browser-native platform for interactively exploring how algorithms work — 138+ algorithms across 15 categories, with step-by-step playback you can scrub like a video.
Highlights
- Snapshot-based execution model — deterministic O(1) seeking, no re-computation
- 8-language code viewer (C, C++, C#, Go, Java, JavaScript, Python, Rust)
- Algorithm-specific visual signatures — AVL balance factors, KMP failure tables, Dijkstra relaxation, and more
- Three-level fallback system so the UI never goes blank, strict CSP and security headers
Stack: React 18 · TypeScript 6 · Vite 5 · TailwindCSS · Vitest Repo: github.com/ezpectus/algo-visualiser · Live: algorithms-laboratory.netlify.app
Algorithms have been a constant thread through my engineering journey — nearly 3,200 problems solved across LeetCode, HackerRank, and Codeforces.
- LeetCode Solutions — C# and C++ implementations
- Codeforces Solutions — C++17 contest archive
- Algorithm Patterns — rare algorithms with architectural breakdowns and complexity analysis
I view algorithmic practice as a way to sharpen engineering thinking and architectural decision-making — not just interview prep.
Led development on university and personal projects — architecture planning, technical decision-making, task decomposition, code reviews, and keeping development standards consistent across a team.
Distributed Systems · RabbitMQ · Advanced PostgreSQL · SaaS Architecture · Backend Scalability · Production System Design
Software Engineer / Backend / Full-Stack roles · Graduate Software Engineer positions · Freelance projects · Open-source collaboration
I enjoy building software that stays maintainable, scalable, and useful long after the first version ships.