Modern energy infrastructure spanning power grids, generation plants, and oil and gas networks has evolved into a massive ecosystem of interconnected, smart physical assets. The operational efficiency of these systems is no longer just a function of localized physics; it is driven by a complex mix of market economics, automated edge-computing algorithms, asset degradation constraints, and human decision-making.
Our division is building the mathematical and algorithmic foundation to understand, predict, and optimize these large-scale systems. We are moving beyond traditional physics-only simulations to build behavioral models. By abstracting physical assets (turbines, compressors, transformers, and switchgear) into interacting, intelligent agents, we simulate and optimize how they will behave when exposed to volatile market dynamics, network constraints, and operational uncertainties. We bridge the gap between hard engineering realities (thermodynamics, fluid dynamics, electromagnetics) and AI-driven decision-making.
The Role
We are seeking a Senior Quantitative Scientist to anchor our mathematical research and development. You do not need to be a domain expert in every industrial sector. Instead, we are looking for a rigorous mathematical thinker who can bring deep expertise from advanced optimization, game theory, or AI/ML, and apply it to complex physical networks. You will collaborate closely with domain engineers (electrical, mechanical, chemical) and systems architects to ensure your mathematical abstractions accurately reflect the constraints of real-world industrial operations.
Key Responsibilities
- Complex Systems Modeling: Design data-driven and algorithmic models that capture the complex, often unpredictable behavior of decentralized physical assets under volatile market conditions and varying environmental scenarios.
- Agent-Based & Behavioral AI: Develop reinforcement learning and multi-agent frameworks that simulate how autonomous industrial devices interact, compete, and cooperate within a shared network or supply chain.
- Economic & Incentive Design: Help design the algorithmic "rules of the game" (automated local markets, resource allocation mechanisms, and scheduling algorithms) that incentivize self-interested assets to operate in ways that support overall system stability and profitability.
- Optimization Under Uncertainty: Build robust optimization models that maximize efficiency and asset longevity even when facing worst-case scenarios, erratic demand distributions, or incomplete data.
- Physics-Aware AI: Collaborate with engineering teams to embed strict physical safety limits (e.g., pressure, temperature, thermal, and voltage constraints governed by PDEs) directly into your data-driven models, ensuring the algorithms never violate fundamental physical laws.
- Thought Leadership: Drive the research agenda for the division, mentoring junior scientists and representing our work in industry forums and academic publications.
Required Qualifications
- Education: PhD in Applied Mathematics, Computer Science, Operations Research, Statistics, Electrical Engineering, or a closely related discipline.
- Experience: 10–12+ years of research or industry experience focused on advanced mathematical modeling, machine learning, and algorithm design.
- Core Expertise: Deep, provable expertise in at least two of the following domains:
- Game Theory, Mechanism Design, or Behavioral Economics
- Advanced Mathematical Optimization (e.g., Stochastic, Robust, or Non-linear programming)
- Reinforcement Learning / Multi-Agent Systems
- Physics-Informed Machine Learning (PINNs) or constrained AI models
- Technical Stack: Fluency in Python and deep learning frameworks (PyTorch, JAX, or TensorFlow). Experience with mathematical solvers (e.g., Gurobi, SCIP, CVXPY) is highly desirable.
Preferred Qualifications
- Domain Exposure: While not strictly required, prior exposure to energy systems, smart grids, algorithmic trading, or complex supply chain networks is a strong plus.
- Research Pedigree: A track record of publishing impactful research in top-tier AI/ML venues (e.g., NeurIPS, ICML, AAAI, AI-Stats) OR premier domain-specific journals (e.g., IEEE Transactions on Power Systems, Smart Grid).
- Cross-Disciplinary Collaboration: Proven ability to translate abstract mathematical concepts into practical tools for engineering and product teams.
