The company is rebuilding how energy is bought and sold, cutting out middlemen and connecting businesses directly with renewable generators, with every transaction traceable and transparent.
It is a technology company working on the energy industry rather than an energy company. The team has 160+ people across operations, data science, engineering, customer work and energy expertise.
The company has raised $94M in total funding, including a $75M Series B in early 2026, and serves over 4,000 UK businesses across 8,500+ sites, with over £75m returned to businesses and generators so far.
The role is Staff Machine Learning Engineer (L4) on the pricing team, working in Rosso, the core system that forecasts prices and volumes, tracks the live portfolio and sets quote fees. You will join the pricing ML team and own key parts of the dynamic pricing system. Direction is already set; within your parts, you choose the bets and prove whether they worked.
Responsibilities
Own key parts of the dynamic pricing system from design through to production operation.
Pick the bets in your area most likely to cut sales adjustments, and show the results.
Make your components observable, guarded and replayable so a bad price is caught before a customer sees it.
Clear your own blockers with engineering, platform and commercial teams.
Own one or more core components, such as models that learn from won and lost deals, the experimentation and measurement layer, or the live guardrails that limit risky price changes.
Compare competing bets in your area by how likely each is to move the metric, and stop the weak ones.
Design experiments and holdouts for low-volume, high-value deals, where a standard A/B test is often not feasible.
Build monitoring, alerting, guardrails and staged releases, including drift detection, catching runaway feedback loops, and holding unexpected price moves for review.
Make every pricing decision your components produce reproducible later, for debugging and for regulators.
Work with the Price Quoting squad and ML Platform so models are production-ready from the start.
Find out why sales teams override prices and turn those reasons into model inputs or new bets.
Explain model behaviour and trade-offs clearly to engineers, product and commercial colleagues so they can act on it.
Redesign the team's recurring pricing workflows, such as feedback triage, daily price review and experiment write-ups, with AI built in, and decide which steps stay human.
Requirements
You write and ship production code as most of your job and want to keep doing so. This is not a people management role or an architecture role where others do the building.
You have owned a live ML system making automated decisions with money involved, such as algorithmic trading, dynamic pricing, marketplaces, ad bidding, credit or betting.
You have handled production failures such as feedback loops, silent drift, bad outputs reaching customers and rollbacks, and changed your practice because of them.
You have strong probabilistic modelling and decision-making under uncertainty, including pricing, conversion or elasticity modelling, and experimentation. You frame problems from first principles and choose the simplest approach that works.
You have run ML as production services on modern cloud infrastructure, such as containerised models behind APIs, event-driven microservices, infrastructure as code and CI/CD on AWS or similar.
Model changes go through the same pipeline, tests and rollback as other code, so deploying one late on a Thursday is routine for you.
You write production-grade Python and design for observability, testing and replay. You work as a peer of software engineers.
You have a record of moving a commercial metric, working with product managers and commercial stakeholders.
You use AI tools daily to multiply your output and know where they fall short.
Energy experience is not required; the company will teach you the market. What is hard to teach quickly is judgement from running a live pricing or trading system.
Nice to have: experience in energy, commodities or other markets where prices, forecasts and risk interact.
Nice to have: contextual bandits, causal inference or Bayesian methods used in production.
Nice to have: stochastic optimisation or operations research.
Nice to have: a PhD in a field such as maths, physics, computer science or operations research.
Conditions
Fully remote, with the team distributed across Europe.
Hiring process takes about 2-3 weeks from first call to offer. Tell the team if you need timeline adjustments.
Step 1: a 30-minute call with the Talent Team about your experience, motivations and the role.
Step 2: a 30-minute behavioural interview with the Rosso GM, who is the hiring manager.
Step 3: a 90-minute technical interview with the team, including a live technical exercise with potential peers.
Step 4: a 45-minute final interview with two cross-functional stakeholders.
The company welcomes applicants of all backgrounds and identities, including people underrepresented in tech and energy, and encourages applying even if you do not meet every requirement.
Benefits
Salary of £115,000, or the equivalent in local currency.
Stock options, giving everyone on the team ownership.
25 days holiday plus public holidays, where you can swap public holidays for ones that matter most to you, and an extra day off for your birthday.
Up to £1,200 (or €1,200) a year to upgrade your remote setup, such as co-working passes and equipment.
Up to £150 (or €150) a month for anything that supports your wellbeing, from therapy to gym memberships to meditation apps.