Senior Machine Learning Engineer

• Join a two-person ML team and a forecasting system making half-hourly demand predictions across diverse targets for multiple restaurant chains. • Work on models that enter a complex environment where key machine and human decisions rest on their predictions, facing feedback loops and high variability. • Take on the core challenge: the system works, so now scale it from a handful of clients to hundreds. • Own client launches end-to-end — validating data, selecting models, running UAT, going live, and monitoring performance afterwards. • Join client calls, build relationships, and understand what actually matters on the ground — not just whether the model is accurate, but whether the kitchen prepped the right amount of food.

• Bring production ML experience, with a track record of deploying systems that handle messy data, fail gracefully, and need monitoring. • Bring strong ML fundamentals, reasoning about trade-offs in practice, explaining the 'why' behind feature and model choices, and making good judgement calls when something unexpected happens. • Bring client-facing deployment experience, having personally owned an ML deployment end-to-end and being comfortable on calls with non-technical stakeholders. • Bring strong Python programming skills, including the ML and scientific Python stack such as numpy and scikit-learn. • Use AI development tools daily (Claude Code, Cursor, Copilot, or equivalent) as your default working mode. • Favour getting something simple in front of a user over complicated and shiny, working with pragmatism, parallelism, and iterative improvement.

• Bring experience in forecasting, time-series, or demand-planning, with an intuitive grasp of lag features, calendar effects, and evaluation integrity. • Have experience with the stack: Python, scikit-learn, MLflow, Docker, and GCP. • Have worked in a small team where ownership is wide and context-switching is normal. • Have taken a machine learning project from business idea through to deployed production system.

• Own client launches end-to-end, from validating data and selecting models to running UAT, going live, and monitoring performance afterwards. • Scale the forecasting system from a handful of clients to hundreds. • Join client calls and build relationships, understanding the realities of what's happening in each client's business. • Work with industry experts to understand what's actually driving the data and the opportunities it creates. • Deliver best-in-industry models while keeping clients happy and their businesses running successfully. • Default to AI tools for ML development as your standard working mode.

Sona is one of the most exciting B2B SaaS companies in the UK right now. The company has raised over $100 million from world-class investors and has grown revenue by 400% in 2023 alone. Two of the three founders, Oli Johnson and Ben Dixon, previously spent six years building Catapult, a gig platform for hospitality and retail, while Steffen Wulff Petersen came from Rocket Internet, where he helped launch Lazada, HelloFresh and Payleven. That domain expertise is a genuine differentiator: Sona is not a generic scheduling tool bolted onto HR. It is built from the ground up for complex, multi-site enterprises with hundreds or thousands of frontline staff. What stands out on Glassdoor (4.6/5 from 26 reviews, 88% recommend, 94% positive business outlook) is how consistently employees praise the team culture, the benefits and the pace of growth. The compensation sub-score is a remarkable 4.9/5. Sona offers genuinely strong benefits for a startup: Bupa health insurance, 25 days holiday plus 10 flexible public holidays (with an extra day per year of service), enhanced parental leave (26 weeks full pay for primary parents, 8 weeks for secondary), share options, pension matching up to 5%, twice-yearly team retreats and a quarterly £200 learning budget. The main things to be aware of: this is a fast-paced startup where priorities shift quickly; some reviews mention long hours and pressure around headcount management; and the team is still relatively small, at around 150 people. But with a fresh $45 million Series B led by N47, US expansion underway, and the recent launch of Forge, an enterprise AI application builder that lets customers create custom software on top of Sona's platform, the company is at a genuinely exciting inflexion point.

• Earn a base salary of £95,000–£110,000 with share options. • Work fully remote within a European timezone. • Take 35 days' annual leave (25 days standard plus 10 flexible public holiday days), with an extra day for every year of service. • Get pension contributions matched up to 5% and comprehensive health insurance. • Benefit from enhanced parental leave and pay, plus a salary sacrifice childcare scheme (workplace nursery). • Claim a co-working space stipend if based outside London. • Join bi-annual, all-expenses-paid team retreats. • Receive the latest MacBook, a home-office equipment budget, and a professional development budget. • Enjoy unlimited free books.

Sona is an AI-native platform for frontline workforce management, built for the largest frontline organisations. Its end-to-end system covers Scheduling, HR, Payroll, and Communications, giving organisations everything they need to staff more intelligently and empower their teams. In under five years, Sona has made a deep impact on the lives of over 100,000 frontline workers, grown the team to 140+, and secured over $100M in funding from notable VCs, including a Series B led by N47 alongside Felicis, Northzone, and Gradient Ventures (Google).