Principal Machine Learning Engineer (Revenue Optimisation)

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.
• (Desirable) Experience with real-time pricing at scale — ride-hailing, food delivery, logistics, or similar environments where latency and portfolio effects matter.
• (Desirable) Familiarity with energy markets, power trading, or portfolio risk management.
• (Desirable) PhD or equivalent research depth in a quantitative discipline — statistics, applied mathematics, operations research, or similar.
• (Desirable) 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.
• (Desirable) Experience with causal inference or reinforcement learning in applied commercial settings.

What the job involves

Rosso is tem's core IP. It's the transaction infrastructure that replaces what a traditional trading desk does — forecasting energy prices and volume, building a real-time picture of the portfolio, optimising the fees placed on every quote, and autonomously managing hedging decisions. All of it running continuously. All of it on the critical path for every deal tem closes.

Machine learning is at the heart of it. Rosso combines forecasting, optimisation and classical ML to process billions of data points and drive thousands of automated decisions a day. Every inference shapes the prices our customers see.

We've proved the concept. tem now serves 2% of the UK market. The next step is building a pricing engine that doesn't just react — one that proactively drives growth by targeting the right customers at the right time, at the right price, while protecting margin and portfolio balance. Then taking that internationally.

We're looking for a Senior Staff Machine Learning Engineer to own pricing ML within Rosso. This is a hands‑on senior IC role with real technical authority — you set the strategy, define the mathematical approach, build the models, and ship them. You work closely with MLOps and software engineers, but you don't wait on them. The hard part of this job is the formulation, not the infrastructure.

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.