Principal MLE – Revenue Optimisation

Role Overview

• Own the technical direction for pricing ML. Define what to build and how. Set the roadmap for the pricing engine as a core piece of tem's IP — and be accountable for its performance.
• Formulate and solve the pricing problem properly. The mathematical foundation doesn't fully exist yet. Your first job is to define it: a dynamic, real-time system that simultaneously optimises for signing probability, portfolio balance, and margin. Choose the right approach — stochastic programming, reinforcement learning, classical ML, or a hybrid — based on the problem, not familiarity.
• Build and ship models end-to-end. Own the modelling and data layer. Write production-grade Python. Architect models with deployment in mind and carry them through to production — you can execute without being blocked by engineering dependencies.
• Solve imbalance problems. Develop probabilistic models to optimise risk management and short-term balancing decisions in a highly dynamic environment.
• Be the voice of pricing ML across the business. Commercial, product, and engineering teams depend on this engine. They need to understand what it's doing and why. You make that happen — clearly, without losing precision.

Requirements

• Deep experience building ML systems for pricing, revenue optimisation, or real-time decision-making — at companies where pricing is the product, not a supporting function. Track record of models that reached production and moved commercial metrics.
• Strong foundation in stochastic optimisation and probabilistic modelling. The judgement to formulate ambiguous business problems mathematically before reaching for a tool.
• First-principles reasoning across methods. You choose between stochastic programming, reinforcement learning, classical ML, or a simple heuristic based on what the problem demands.
• The engineering depth to match your modelling. Production-grade Python, high bar for code quality, and the ability to carry models from formulation to deployment without being blocked.
• Senior technical leadership. A track record of setting direction for a significant technical area, influencing cross-functional teams, and translating complex model behaviour into clear terms for commercial, product, and engineering stakeholders — so decisions are understood and acted on.
• Experience with real-time pricing at scale — ride-hailing, food delivery, logistics, or similar environments where latency and portfolio effects matter.
• Familiarity with energy markets, power trading, or portfolio risk management.
• PhD or equivalent research depth in a quantitative discipline — statistics, applied mathematics, operations research, or similar.
• Ability to reason about trade-offs between optimisation solvers (Gurobi etc.) and gradient-based methods (PyTorch etc.), and the judgement to know when to reach for each.
• Experience with causal inference or reinforcement learning in applied commercial settings.

Tech Stack

• Python
• PyTorch

Benefits

• Offers Equity