Robotics student at the Technical University of Moldova, focused on backend and full-stack software engineering, AI systems integration, and developer tooling with Python and TypeScript.

Chișinău, Moldova · LinkedIn · GitHub

Interested in backend/full-stack internships, junior software engineering opportunities, and remote or international technical collaboration.

I focus on explicit contracts, deterministic validation, provenance, and reliable handling of uncertain or untrusted inputs.

A timetable application with geometry-based PDF parsing, validated updates, and recovery from accepted schedules. My work includes source acquisition and publisher tooling.

TypeScript · Next.js · React · PostgreSQL · Open timetable

Not an official university service.

AI-assisted purchase-order interpretation with deterministic contract reconciliation, source-grounded corrections, and human approval. A local MVP with explicit live-provider and replay modes; AI cannot approve orders.

Team Lead; contributions include provider/API integration, deterministic discrepancy checks, and reproducible evaluation.

Python · FastAPI · Pydantic · SQLite · JavaScript · Case study / local demo

Python CLI and library for structured source context: filtering, token budgets, optional redaction, AST skeletons, and omission markers.

Python · CLI · AST processing · Unit tests

Packaging decision-support prototype with deterministic calculations, explicit unknowns, and qualification gates. Modeled inputs and calculated results do not establish physical qualification or verified savings.

Team Lead; contributions include domain validation, eligibility safeguards, and Selection integration/demo checks.

Python · FastAPI · React/TypeScript · Case study · Live static demo

Browser/SVG map connecting planned timetable events to explicitly identified rooms, with separate real-data and synthetic-demo modes. Coverage is incomplete; planned classes are not measured occupancy. An independent university-related prototype.

JavaScript · SVG · API validation · Open map

• Smart Harvest, a team training prototype using deterministic baselines and ranked batch assessments, not a production ML model. Python/FastAPI · React/TypeScript. Contributions: contract hardening and pinned-baseline serving.
• GigaHack preparation: dated research, engineering playbooks, and collaboration workflows.

Python, TypeScript/JavaScript, FastAPI, Pydantic, React/Next.js, SQLite/PostgreSQL, structured LLM integration, and automated testing.