Eugen Soloviov
Trading-systems engineer
Trading-systems engineer building bots since 2017: cross-exchange arbitrage (connected up to 30 venues), cointegration-based pairs arbitrage across spot and futures, scalping, news and sentiment-driven strategies, trend algorithms, and portfolio management and balancing algorithms. Also builds sub-millisecond order execution, big-data warehouses, backtesting engines, AI agents, and trading interfaces (incl. open-source profitmaker.cc). Stack: JS/TS, Python, Rust/Zig/Go, DevOps, backend, frontend, architecture.
Articoli
Irregular Time in Tick Models: Continuous-Time Encodings vs. Plain Positional Embeddings
Sequence models fed tick data still assume regular spacing. Three ways to tell a Transformer when a tick actually happened — learnable-timescale continuous encoding, ODE-RNN latent state, and delta_t as a plain feature — and the ablation that decides between them.
Metodi di controllo sintetico per la valutazione delle strategie di trading
Questa serie ha valutato il percorso di selezione fino a un falso limite: il DSR valuta il vincitore, il PBO valuta la ricerca. Nessuno dei due tocca la confusione: la strategia che ha fatto soldi perché la volatilità è raddoppiata nella settimana in cui l'hai implementata. Il metodo di controllo sintetico crea un controfattuale ponderato da un pool di donatori di strumenti intatti e fornisce un criterio di falsificazione e un valore p placebo.
Punteggio delle previsioni probabilistiche: CRPS, calibrazione PIT e DeepAR
Come valutare onestamente una distribuzione predittiva: CRPS come regola di punteggio adeguata, l'istogramma PIT come diagnostica di calibrazione e campionamento DeepAR in GluonTS.
Reti neurali informate dalla fisica per i prezzi delle opzioni
Mettere le PDE dei prezzi Heston, a confine libero e con diffusione a salto in una perdita di rete neurale: il residuo del prezzo logaritmico, il trucco autogradante misto-parziale e ciò che deve ancora essere misurato.
PCMCI: Causal Discovery in Multivariate Crypto Time Series
How PCMCI's two-stage MCI test recovers directed causal links between crypto assets where correlation and bivariate Granger cannot — the construction, the tigramite pipeline, and the real-data study it still needs.
Order Flow Imbalance: The Cont-Kukanov-Stoikov Event Decomposition
Turning raw book updates into a signed flow quantity: the CKS event decomposition, multi-level OFI with PCA reduction, and Lee-Ready trade classification — plus an honest accounting of what the headline R-squared actually measures.
Neural ODEs: Does Continuous Time Beat a Delta-Time Feature?
Neural ODEs, Neural SDEs and continuous normalizing flows for irregularly-sampled market data — and the one ablation that decides whether continuous dynamics are worth their solver cost.
Multi-Task Learning for Simultaneous Price, Volume, and Volatility Prediction
Does jointly predicting return, volume, and volatility actually help? Measuring loss-balancing schemes and diagnosing negative transfer through gradient cosine similarity — with a classical baseline and purged walk-forward folds.
Model Pruning for Low-Latency Trading Inference
Magnitude and structured pruning, the Lottery Ticket Hypothesis, movement pruning, distillation and 2:4 sparsity — the methods behind shrinking a trading model, and what still has to be measured before any of it ships.
Updating the Volume Curve Intraday: Does Adaptive Forecasting Actually Help?
Our VWAP article shipped a static, weekly-refit volume curve and called the forecaster the weakest link. This is the follow-up: a Bayesian intraday updater run against the same 500-parent BTCUSDT harness, with the IS delta conditioned on realized curve error.
Koopman Operators and DMD: Do Market Modes Survive Out-of-Sample?
Dynamic Mode Decomposition fits a linear operator to nonlinear market dynamics. The only question that matters: do the fitted modes persist from one window to the next, and does the rolling spectral radius lead realised volatility? Here is the measurement protocol and the code to run it.
Knowledge Distillation: Compressing Trading Models for Low-Latency Deployment
The blog's standing answer to the accuracy-vs-latency tension is a two-stage fast/slow split. Distillation is a different answer: train one small model to mimic the ensemble. The KD loss, temperature, born-again nets, early exits for a variable latency budget, and the distill-to-FPGA pipeline — plus the measurements that would decide whether it beats the two-stage split.