Data Scientist – Quantitative Trading

In line with TotalEnergies’ objective of achieving carbon neutrality by 2050, the company is rapidly expanding its global presence across the entire power value chain, with a particular focus on the fast‑growing sectors of renewables and flexible generation assets. To fully capture the value of these power assets in spot markets, TotalEnergies Gas & Power aims to participate at every stage of the value chain—from curve trading to Day‑Ahead optimization, Intraday activity, and ancillary services. Short‑term power optimization is built on a U.S. framework that aggregates generation assets (CCGT, wind, solar), flexibilities (DSR, batteries), and the power‑supply portfolio to maximize value creation for the company.

The job holder is part of the Algorithmic Trading team based in Houston and collaborates closely with sister teams in Geneva. The team operates within an uncertain and highly volatile energy‑commodity trading environment influenced by global economic conditions, energy‑resource development, political and legislative changes, climate‑related factors, and evolving energy demand. It is a competitive landscape with significant pressure, time‑critical activities, and a high‑stress working environment.

Operations run automatically seven days a week with daily submissions. Onsite presence is required five days per week, with periodic short‑term remote oversight on weekends and holidays.

The role leads the team’s quantitative and modeling efforts at the forefront of data science in the energy sector. You will be expected to apply advanced modeling techniques to support strategic decision‑making and contribute directly to P&L generation. You will also provide data‑science expertise across the team, offering support on a wide range of modeling and optimization topics.

• Purpose: The position sits within the Modelling and Analysis for Trading Strategy (MATS) department, which leads the end‑to‑end design and development of trading strategies, quantitative models, advanced analytical and AI tools, and market research that support optimized decision‑making across all TGP trading desks. Positioned at the strategic intersection of Trading, Risk Management, and IT, the team ensures agility, robustness, and continuous innovation in the algorithms that drive TGP’s market strategy.
• Objectives: Contribute to the growth of the Algo‑trading activity by developing forecasting models, decision‑making algorithms, and positioning optimizations. Build virtual DA‑RT strategies and intraday proprietary trading strategies within the established risk‑policy framework and compliance rules for U.S. nodal power markets.
• Impact: The role directly influences TGP’s trading performance by supporting long‑term P&L growth through scalable trading strategies and durable analytical infrastructure. It requires strong technical skills, sound risk awareness, critical thinking, and a collaborative mindset. This is a strategic position that strengthens TotalEnergies Gas & Power’s presence in U.S. short‑term power markets.
• Location: Based in Houston.

Key Tasks:

• Develop trading models using the MATS libraries, covering the full lifecycle from data ingestion to live result delivery, including feature engineering, data optimization, model tuning, risk and positioning models, performance monitoring, and probability meta‑optimization.
• Create new indicators and features to enhance machine‑learning and quantitative models used in algorithmic trading.
• Monitor model performance in both live and paper‑trading environments.
• Oversee production deployment of models in collaboration with squad members and cross‑functional MATS teams.
• Improve algorithmic‑trading strategies through research on U.S. nodal power markets.
• Analyze large datasets to generate meaningful insights that support strategy development.
• Coordinate with team members to ensure high‑quality back‑testing, validation, and methodological robustness.

• Talented and highly motivated individual with a strong technical or engineering background, holding a BEng, MSc, or PhD in a quantitative field such as Mathematics, Physics, Machine Learning, Optimization, or Computer Science.
• One to three years of experience deploying machine‑learning models in production (delivering live forecasts) and/or formulating and solving bet‑sizing and positioning problems as stochastic optimization under uncertainty.
• Proficiency in Python, MongoDB, and Shell scripting, with a solid understanding of software architecture and SQL.
• Familiarity with Docker and DevOps/MLOps practices, including CI/CD workflows.
• In‑depth knowledge of machine‑learning techniques, including supervised and unsupervised modeling.
• Strong understanding of quantitative bet‑sizing methods such as utility‑based approaches, Kelly‑fraction techniques, and risk‑constrained optimization.
• Practical experience with time‑series analysis, particularly in the context of energy markets and trading signals.
• Ability to analyze complex systems and identify indirect correlations across diverse signals.
• Strong project‑management capabilities with clear communication and effective reporting skills.

Additional and advantageous qualifications:

• Broad understanding of energy and commodities markets, including supply‑side dynamics.
• Experience with project‑management tools such as Wrike, Jira, or Linear.