Eugen Soloviov
Trading-systems engineer
Ingeniero de sistemas de trading que construye bots desde 2017: arbitraje entre exchanges (conectado hasta 30 plataformas), arbitraje de pares basado en cointegración entre spot y futuros, scalping, estrategias basadas en noticias y sentimiento, algoritmos de tendencia y algoritmos de gestión y balanceo de carteras. También construye ejecución de órdenes en submilisegundos, almacenes de big data, motores de backtesting, agentes de IA e interfaces de trading (incluido el open-source profitmaker.cc). Stack: JS/TS, Python, Rust/Zig/Go, DevOps, backend, frontend, arquitectura.
Articles
Digital Fingerprint of a Trader: How to Identify a Market Maker by Their Order Book Behavior
Every algorithm leaves a unique fingerprint. Learn to read it — and you will know who is on the other side of your trade.
PnL by Active Time: The Metric That Changes Strategy Rankings
Why raw annual PnL is a poor metric for comparing strategies with different trading time. How to calculate effective return, why you need fill_efficiency, and why a strategy with 27% PnL can outperform one with 300%.
Adaptive Drill-Down: Backtest with Variable Granularity from Minutes to Raw Trades
How adaptive data granularity speeds up backtests and saves storage: drill-down from 1m to 1s, 100ms, and raw trades only where price moved significantly or volume spiked, not across the entire historical series.
Aggregated Parquet Cache: How to Speed Up Multi-Timeframe Backtests by Hundreds of Times
How to precompute timeframes and indicators from minute candles, save them to parquet, and use them for mass strategy testing without redundant recalculations.
Walk-Forward Optimization: The Only Honest Strategy Test
Why a single train/test split does not protect against overfitting, how walk-forward optimization systematically verifies parameter robustness, and why a strategy with +3342% PnL@ML on 21 parameters is a ticking time bomb without WFO.
Signal Correlation: How Many Pairs to Monitor
Why 10 crypto pairs don't provide 10x diversification, how to calculate effective_N via correlation_factor, and how many pairs you really need to monitor for 80-90% orchestrator slot utilization.
Polars vs Pandas for Algotrading: Benchmarks on Real Data
Detailed comparison of Polars and Pandas on algotrading tasks: benchmarks for filtering, aggregation, rolling signal computations, I/O, and memory consumption. Hybrid Polars + Numba architecture for maximum backtest performance.
Plateau Analysis: How to Distinguish a Robust Optimum from Overfitting
Why finding the best strategy parameters is only half the work. How to visually and quantitatively distinguish a stable plateau from a fragile peak, and why Optuna contour plots are a mandatory step before launching an optimized strategy into production.
Coordinate Descent vs Bayesian Optimization: Which Finds Better Parameters
Why exhaustive search is impossible for 12+ parameters, how coordinate descent misses interactions, and how Optuna with a TPE sampler finds in 500 iterations what OAT cannot find in 96. Practical code examples, sampler comparison, and multi-objective optimization.
Multi-Symbol Validation: Test Your Strategy on All Pairs
Why a strategy optimized on ETHUSDT may fail on altcoins. How to properly test across pair groups (blue chips, large caps, shitcoins) and what cross-symbol robustness score to consider sufficient.
Funding Rates Kill Your Leverage: Why PnL×50x Is a Fiction
How funding rates on Binance/Bybit turn beautiful high-leverage backtest results into guaranteed losses. Formulas, recalculation of real strategies, and the maximum leverage at which funding does not eat into profits.
Cascade Strategies: Priority Execution with Fallback Filling
Finale of the 'Backtests Without Illusions' series. How to build an orchestrator from N strategies x M pairs, implement cascade mode with priority and fallback filling, choose dual_size, and why strategy portfolios cannot be backtested by summing PnL.