I’m a passionate software creator specializing in designing and building complex, efficient, and scalable systems. My work focuses on bridging high-level architectural vision with low-level implementation to deliver solutions that are not only technically elegant but also pragmatic and effective.
- đź” Currently exploring the performance frontiers in Rust and GraalVM Native Image.
- 🌱 Honing my skills in declarative system management using NixOS.
- đź’¬ Happy to discuss software architecture, compiler theory, distributed systems, and AI modeling.
As a developer, I navigate between different paradigms and ecosystems—from low-level system programming in Rust and C, to building robust enterprise applications in Java (Spring Boot), to crafting dynamic interfaces in TypeScript/React. I’m fascinated by performance optimization, automation, and creating reproducible development environments, reflected in my projects using GraalVM, Docker, and NixOS. I believe the best software emerges at the intersection of deep computer science fundamentals and a practical approach to solving real-world problems.
Programming Languages
|
Backend & Databases
|
Frontend
|
DevOps, Systems & Tools
|
Below is a detailed overview of my most important work. Projects are grouped thematically to showcase the breadth of my skills.
In this section, I present deep technical explorations that go beyond standard implementation. These are case studies and architectural specifications demonstrating the ability to design complex systems from scratch.
Project Nexus: Specification for a Unified Programming Language
> **Project Goal:** Design and specify a new programming language (`.nx`) and its ecosystem. The aim is to create a successor that synthesizes the performance of C, the productivity of Python, and the asynchronicity of JavaScript into a single coherent paradigm. > > **Key Architectural Concepts:** > - **Unified Intermediate Representation (UIR):** The core system—a common intermediate representation into which code from various languages is compiled, enabling unprecedented optimizations. > - **Library Ingestor (`nexlink`):** A tool for transpiling existing libraries (C, Python, JS) into the native `.nexlib` format, eliminating the need for FFI (Foreign Function Interface). > - **Polyglot Optimizer:** An advanced optimizer capable of cross-language transformations, such as inlining a Python function directly into a Nexus loop. > - **Dual Memory Model:** Default safe memory management (ARC + GC) with the ability to switch to a `perf {}` block with manual allocation and borrow-checking inspired by Rust. > > **Project Value:** Demonstrates deep understanding of **compiler theory, language design, operating systems, and complex software architecture**.Project Phoenix: Hybrid Desktop Application Architecture
> **Project Goal:** Design a hybrid desktop application architecture that combines **instant startup and low memory usage** (characteristic of native apps) with **the speed and flexibility of web UI development**. > > **Key Architectural Concepts:** > - **Native Backend (AOT Compilation):** Backend in **Spring Boot** compiled to a native binary using **GraalVM Native Image**, ensuring startup in a fraction of a second. > - **Java Frontend:** User interface written in Java, then compiled to high-performance JavaScript/WebAssembly using tools like `J2CL`. > - **Desktop Shell:** Minimalist **JavaFX** application with a `WebView` component that renders the web frontend, running on a standard JVM. > - **Strict API Contract:** Shared Maven module with DTOs, ensuring type safety and serving as a formal contract between client and server. > > **Project Value:** Shows the ability to design **complex, multi-module systems**, knowledge of advanced compilation techniques, and making informed architectural trade-offs.Architectural Case Studies (Startups as Technical Ideas)
> These projects are technical analyses of business problems, presented as ready-made architectural concepts for hypothetical startups. > > 1. **Intellexa - Market Intelligence from AI Conversations:** > - **Problem:** Companies have millions of AI chatbot logs ("dark data") that are an untapped resource. > - **Technical Solution:** Three-tier SaaS platform (B2B). **Tier 1 (Ignite):** Analytical dashboard with PII anonymization and topic clustering. **Tier 2 (Accelerate):** Adds API for insights and benchmarking. **Tier 3 (Apex):** Introduces Fine-Tuning as a Service (FaaS) and dedicated data strategist support. Architecture based on microservices, stream processing, and NLP models. > > 2. **Integrit Labs - Integration as a Service:** > - **Problem:** Engineers waste valuable time integrating and maintaining external SDKs (Stripe, Twilio, Scandit), instead of focusing on product development. > - **Technical Solution:** Integration services company evolving into a platform. **Phase 1:** Expert services (fixed-price). **Phase 2:** Premium WordPress plugin (`Integrit Connect`). **Phase 3:** Creation of `Integrit API`—a unified "API to API" that becomes the main SaaS product. > > 3. **LearnSphere Academy - EdTech Platform:** > - **Problem:** Need for a professional, engaging educational platform. > - **Technical Solution:** Landing page design and initial system architecture. Frontend in **Next.js** (for SEO and performance), backend as a **Headless CMS** (e.g., Strapi) for content management (courses, tutors) and API for handling registrations and consultations. Future integration with **SphereLabs AI**—a custom LLM module for personalized learning. > > **Project Value:** Demonstrates the ability to **translate business needs into concrete technical solutions** and think about scalability and monetization.Self-Hosted AI Chat (Rust & React)
> **Description:** Fully functional, locally hosted AI chat application. Backend written in **Rust (Actix Web)** uses the `llama-cpp-2` library for inference on LLM models (GGUF format) without external APIs. Frontend built in **React** and **TypeScript** provides a modern, responsive interface. > > **Key Features:** > - No dependency on external APIs—100% privacy. > - Conversation context management. > - High performance thanks to native backend. > > **Technologies:** `Rust`, `Actix Web`, `llama.cpp`, `React`, `TypeScript`, `Vite`.Pizza Store - E-commerce Application (SPA)
> **Description:** Modern Single-Page Application (SPA) simulating a pizza store. Implemented dynamic product search and filtering (`lodash.debounce`), advanced sorting, cart management, and **PayPal** payment integration. > > **Key Features:** > - State management with **Redux Toolkit**. > - State synchronization with URL and `localStorage`. > - Loading optimization via Code Splitting (`React.lazy`). > > **Technologies:** `React`, `TypeScript`, `Redux Toolkit`, `SCSS`, `PayPal API`.Data Dashboard - Data Visualization Platform
> **Description:** Full-stack application for collecting, processing, and visualizing personal data from various sources. Backend in **Python** periodically fetches and aggregates data. Frontend in **Next.js** delivers interactive charts and filters for data exploration. > > **Key Features:** > - Automated backend tasks. > - Server-side rendering (SSR) for high performance. > > **Technologies:** `Python`, `Next.js`, `React`, `Data Visualization`.Nixrach: Declarative NixOS System Configuration
> **Description:** Comprehensive, fully reproducible **NixOS** system configuration using **Flakes** and **Home-Manager**. Manages the entire development environment—from kernel to window manager (**Hyprland**), applications, aliases, and scripts. > > **Key Features:** > - "Infrastructure as Code" applied to the workstation. > - Modular structure for different hosts (desktop, laptop). > - Full automation and reproducibility. > > **Technologies:** `Nix`, `NixOS`, `Flakes`, `Home-Manager`, `Hyprland`, `Waybar`.envdev: Developer Environment Manager
> **Description:** Native CLI tool (Java + GraalVM) orchestrating a containerized development environment based on **RHEL**. Automatically manages configuration and lifecycle of containers (developer tools + web-based VS Code), providing isolated and reproducible work environments. > > **Key Features:** > - Instant startup thanks to native compilation. > - Dual-container architecture (tools + IDE). > - Full automation via `docker-compose`. > > **Technologies:** `Java`, `GraalVM Native Image`, `Docker`, `Docker Compose`, `Picocli`.2D Real-Time Battle Simulator
> **Description:** Interactive 2D battle simulator with autonomous AI units that independently seek enemies, engage in combat, and react to battlefield conditions. The project uses object-oriented architecture and **HTML5 Canvas API** for real-time rendering. > > **Key Features:** > - Intelligent agent behavior (targeting, pursuit). > - Combat system based on HP and armor. > - Dynamic UI with real-time updates. > > **Technologies:** `TypeScript`, `HTML5 Canvas API`, `Object-Oriented Programming`.Mind-Simulator: Competitive Learning System
> **Description:** **Java/Spring Boot** application simulating an environment where simple, "generic" learning cells compete for data processing. Cells that successfully process data strengthen and develop specializations, demonstrating emergence and system self-organization. > > **Key Features:** > - Competition for resources (data) and specialization. > - Evolutionary cycle (metabolism, death, birth). > - Agent-based architecture. > > **Technologies:** `Java`, `Spring Boot`, `Agent-Based Modeling`.- Collection of 15 Pure JavaScript Projects: Small applications demonstrating in-depth knowledge of fundamental JavaScript mechanisms, DOM manipulation, and event handling without frameworks.
- Telegram Bot in Java: Simple Telegram bot written in Java using the Spring framework, demonstrating integration with external APIs.
- Rust CLI Utility Suite: Collection of command-line tools in Rust, including programs for secure file copying and CSV processing.
- StructUI: Lightweight, class-based UI library written in TypeScript and run with
bun.js.
I’m always open to new challenges and interesting projects. If my work interests you, feel free to reach out!