Віталій

Vitalii Kalinichenko
Junior .NET / Full-Stack Developer
Cherkasy, Ukraine · Remote (EU/US time-zone overlap available) · github.com/VItaly0117 · [відкрити контакти](див. вище в блоці «контактна інформація»)

SUMMARY
Applied physics student (Cherkasy National University) completing an intensive .NET development track alongside my
degree. I build working systems end-to-end — two self-directed algorithmic trading projects (below) with real test coverage
and real paper-trading validation, not tutorials. Comfortable working AI-augmented: I use modern AI coding tools
deliberately, the same way most product teams do in 2026, to move faster without cutting corners on architecture or
correctness.

PROJECTS

Cerberus — Statistical Arbitrage Bot for Polymarket | github.com/VItaly0117/cerberus
Python, asyncio, WebSockets, SQLite, Telegram Bot API
● Built a multi-sprint trading system: market discovery, real-time WebSocket order-book watcher, fee model, risk
manager, and FOK/FAK order executor — 106 automated tests.
● Ran a 72-hour live paper-trading validation; diagnosed and fixed critical bugs (fee-formula error, NULL edge
values, misconfigured attempt caps) that were masking the strategy's true performance.
● Confirmed net edge in the -0.9% to -1.4% range post-fix — an honest negative result that validated the taker-
arbitrage approach was structurally unprofitable at current fee levels, and documented why.
● Engineering constraints self-imposed for correctness: Decimal-only arithmetic (no float drift), sequential leg
execution, DRY_RUN_MODE safety default.
Project Sentinel — Quant Trading Bot for Bybit | github.com/VItaly0117/project-sentinel
Python, XGBoost, Docker, PostgreSQL, FastAPI
● Signal-generation and mean-reversion (z-score) trading bot with a machine-learning confidence layer (XGBoost),
containerized with Docker and backed by PostgreSQL — 64 automated tests.
● Investigated and corrected threshold configuration that had been loosened to force trade activity, restoring honest
signal-confidence and z-score thresholds before evaluating results — prioritizing a truthful baseline over an inflated
backtest.

TECHNICAL SKILLS
● Languages: C#, Python, TypeScript, SQL
● .NET: .NET 8, ASP.NET Core, Entity Framework Core, xUnit
● Architecture: Clean Architecture, CQRS, DDD principles
● Frontend: React, TypeScript
● Data / Infra: PostgreSQL, Docker, GitHub Actions (CI/CD)
● Practice: git workflows, automated testing, AI-assisted development (Claude Code and equivalent tooling) as part of
a modern engineering workflow

EDUCATION
Cherkasy National University — B.Sc. Applied Physics (in progress, 2nd year)
IT Academy — .NET Development Track (intensive, concurrent with degree)

LANGUAGES
● Ukrainian / Russian — native
● English — professional working proficiency (technical reading/writing, spoken communication)