Senior Quantitative Researcher (Time Series ML)

We are a prop-trading firm that blends the agility of a startup with the capabilities of a high-performing fund. We build advanced, data-driven trading strategies across asset classes, and foster a culture where ideas matter, ownership is encouraged, and every team member can unlock their full potential.

Senior Quantitative Researcher with deep expertise in machine learning for time series, strong quantitative intuition, and hands-on experience developing models that drive real-world trading. Experience in options trading (commodities/metals/energy/crypto) is a significant plus.

• Develop and deploy state-of-the-art ML models for forecasting complex, high-dimensional, non-stationary time series (market microstructure, macro data, alternative signals).
• Build full ML pipelines from scratch: data ingestion, feature engineering, modeling, calibration, monitoring.
• Design advanced validation frameworks for non-IID and regime-shifting datasets.
• Work with large-scale, multi-modal datasets (tick, satellite, transactional, scraped, text).
• Formulate, test, and refine investment hypotheses within risk constraints.
• Build and enhance factor models, risk models, and return/volatility forecasting models across asset classes.
• Participate in portfolio optimization, accounting for transaction costs, market impact, and constraints.
• Conduct attribution analysis, stress testing, scenario modeling, and prepare insights for the Investment Committee.
• Develop advanced options strategies, volatility structures, and hedging frameworks in commodities and alternative assets.
• Analyze pricing anomalies, build volatility models (SABR, Heston, local vol), and construct bespoke exposures.
• Contribute to internal options analytics libraries (Greeks, volatility surface, PnL attribution, scenario analysis).

• 7+ years in quantitative research, systematic trading, or ML-driven modeling (mix of industry + academia is ideal).
• Publications in top AI venues (NeurIPS, ICLR, ICML) are a strong plus.
• Demonstrated experience building models that forecast market, macro, or alternative data signals.
• Proven involvement in institutional investment processes (risk, IC meetings, compliance).
• Track record of deploying live strategies or predictive models that delivered measurable performance.
• (Optional) 5–7+ years in commodities/options trading with exposure to metals/energy or alternative assets.

• Expertise in cutting-edge deep learning for time series, including State Space Models (S4/S5, Hyena, Mega), advanced Transformers (iTransformer, TimesNet, TimeGPT-style models), neural SDE/ODE architectures for high-frequency data, diffusion-based forecasting models, etc.
• Strong statistical foundations: regime-switching models, copulas, volatility forecasting.
• Experience with multi-modal learning (time series + tabular + text).
• Proficiency in Python, PyTorch, HuggingFace, DVC, Docker; C++/Rust is a plus.
• Ability to design custom validation strategies for non-IID datasets.
• Understanding of portfolio optimization with linear & nonlinear constraints.
• Master’s or PhD in Physics, Mathematics, CS, or another quantitative discipline.

• Russian, English.

• Knowledge of options pricing, stochastic volatility models, and hedging techniques.
• Experience with ML/DL/RL applied to trading.
• Ability to communicate complex ideas to both technical and non-technical audiences.

• Culture of innovation — a genuinely open, research-driven environment where curiosity is rewarded and your ideas directly shape real trading strategies.
• True flexibility — work from anywhere; we care about outcomes, not where or when you sit at your desk.
• High autonomy & ownership — no micromanagement, no bureaucracy. You get full responsibility over your research direction, models, and production impact.
• Startup agility, Fund resources — fast decision-making, minimal red tape, and access to the data, compute, and infrastructure you need to run serious research.
• Massive data advantage — work with a uniquely rich multi-modal dataset (order log, options chains, satellite data, alt-data, Bloomberg, proprietary feeds).
• Top-tier equipment — choose the hardware/software setup that makes you most productive.
• Well-being support — 35 days of vacation, 100% paid sick leave, and access to a corporate psychologist.
• Real career growth — shape research culture, lead initiatives, and influence long-term strategy directions.