Full-Stack Engineer | AI Systems | EdTech Platforms
I build production systems where AI solves real business workflows: document pipelines, localization engines, LMS ecosystems, and automation services.
- End-to-end AI features in web products, from prototype to production.
- Backend architecture for SaaS and EdTech platforms with clear domain boundaries.
- Automation pipelines for high-volume content, docs, and messaging workflows.
- Research-grade simulation tooling (FEM, numerical modeling, and performance optimization).
- Core: Python, TypeScript, C++
- Web/API: Next.js, React, FastAPI, Node.js
- AI/Data: PyTorch, Pandas, OpenCV, LLM agents
- Simulation: Ansys Maxwell/Electronics, Abaqus, CUDA
- Infra: Docker, PostgreSQL, PocketBase, Railway
| Project | Scope | Stack |
|---|---|---|
| Autotables | AI-driven localization and document generation for automotive specs, including terminology-safe translation pipelines and PDF output automation. | Next.js Python PocketBase |
| EduVenture World | EdTech ecosystem that combines LMS features (video lessons, flashcards, progress tracking) with a service marketplace for study programs. | Next.js PostgreSQL Railway |
| PhD Research (Computational Physics) | Electrical conductivity modeling for 3D-printed Graphene/SiC scaffolds, including current density and spacing impact analysis. | Ansys CUDA NumPy |
| Telegram Automation Ecosystem | Bot suite for SLA auto-replies and Mini App integrations, designed for event-driven and concurrent processing. | Aiogram Python Webhooks |
- Define the target business metric and hard constraints.
- Build a thin vertical slice and validate it with real usage.
- Instrument performance, quality, and failure modes.
- Scale architecture only after measurable proof.
