Syune
Shakaryan

Started in internal automation tooling; grew into backtesting engine development, translating manual trading rules into rigorous, bias-free simulations.

• •Inherited an undocumented backtesting engine and rebuilt its core validation logic to eliminate lookahead bias (strict point-in-time data access) and survivorship bias (sourced historical data including delisted assets).
• •Translated strategy rules from a manual trader into a working backtesting implementation, proposing and confirming rule refinements to close edge cases during implementation.
• •Ran the engine across 3 years of hourly Binance futures data spanning 549 assets (6.2M+ bars), backtesting and evaluating 4–5 trading strategies.
• •Built a systematic parameter-sweep evaluation framework, tracking Sharpe ratio, max drawdown, win rate, and profit factor to separate statistically robust configurations from ones that merely looked good on a single run.
• •Identified and communicated a key finding to the team: high win-rate configurations were catastrophically unprofitable, while lower win-rate configurations with better risk/reward ratios substantially outperformed — reshaping how the team evaluated strategies going forward.
• •Built a Telegram bot handling subscription management for the automated trading service — tracked subscriptions, generated and sent performance summary graphs, implemented a referral system, and integrated with an internal CRM.
• •Built internal tooling (SQL, Streamlit, PyQt6) for operational data tracking.
• •Managed the Linux servers running the company's live trading bots (built by the backend infrastructure team): deployment, monitoring, uptime.