Machine Learning Engineer

Machine Learning Engineer | AI Start-Up | London | £100–120k + Equity The Company Our client is a well-funded, early-stage AI start-up developing advanced machine learning technology to solve complex, real-world challenges in the energy and natural resources sector. Their technology combines sophisticated statistical models with real-world data to help organisations make better decisions around major infrastructure projects, reducing costs, improving efficiency and supporting the energy transition. Backed by leading investors, the business has already secured major commercial customers and is now scaling its technology and engineering capabilities. With a highly technical team of ML researchers, software engineers and domain specialists, they're looking for an experienced Machine Learning Engineer to help take their core models to the next level. Why This Role Stands Out - Work on complex probabilistic and graph-based ML models with genuine real-world applications - Focus on improving and optimising existing models, rather than just deploying them or developing new ones from scratch - Work directly with the CTO and a small team of highly experienced ML researchers and engineers - Take ownership of model performance, accuracy, reliability and computational efficiency - Join an early-stage, well-funded business with established commercial traction The Role As a Machine Learning Engineer, you'll be responsible for improving, optimising and productionising the sophisticated ML models at the core of the company's technology. This isn't a traditional MLOps role or a pure research position. The business already has researchers developing new modelling techniques and an MLOps Engineer responsible for infrastructure and deployment. Your focus will be on understanding the underlying mathematics of existing models and making them faster, more accurate, robust and reliable in production. You'll work closely with the CTO and ML research team, taking ownership of model performance and ensuring the technology can scale across increasingly complex real-world datasets. Your work will feed directly into live customer projects, with the opportunity to see the impact of your improvements in a commercial environment. What You'll Be Working On - Improving the accuracy, reliability and statistical performance of existing machine learning models - Understanding and improving the mathematical foundations of complex probabilistic and graph-based models - Optimising model performance, inference speed and computational efficiency across large-scale datasets - Refactoring, debugging and hardening Python and PyTorch model code for production - Improving uncertainty calibration and model robustness when working with complex, real-world data - Working with ML researchers to turn experimental approaches into reliable, maintainable production implementations - Identifying performance bottlenecks and implementing improvements to model efficiency - Developing rigorous testing and evaluation processes to validate model improvements - Collaborating with the CTO and wider engineering team to influence the technical direction of the product Your Experience - We're looking for a hands-on Machine Learning Engineer with strong mathematical and statistical foundations who enjoys understanding how models work and finding ways to improve them. - You'll ideally have experience in a start-up or scale-up environment, where you've taken ownership of ML models and been responsible for their performance in production. Essential Experience - Strong commercial experience developing, improving and maintaining production machine learning models - Excellent Python and PyTorch engineering skills - Strong understanding of machine learning mathematics, statistics and model evaluation - Experience improving model performance beyond basic hyperparameter tuning - Ability to write clean, well-tested and maintainable ML code - Experience working with complex datasets and optimising models for production environments - Comfortable taking ownership of technical challenges and working collaboratively with researchers and engineers - Strong communication skills, with the ability to explain complex modelling concepts clearly Nice to Have - Experience with Graph Neural Networks (GNNs), graph propagation or PyTorch Geometric - Exposure to probabilistic ML, Bayesian modelling or uncertainty quantification - Experience with numerical optimisation, sparse linear algebra or computational performance profiling - Background working with physical-world, geospatial, environmental, scientific or sensor-based data - Experience taking research models into commercial production - Familiarity with AWS, MLflow, Weights & Biases or similar tooling - Master's degree or PhD in Machine Learning, Statistics, Mathematics or a related discipline Experience in the energy or natural resources sector isn't required. We're particularly interested in engineers who have worked with complex, real-world datasets and can demonstrate strong model development and optimisation skills. Why Join? - £100–120k salary plus meaningful equity in an early-stage, well-backed AI business - Opportunity to work on sophisticated ML models solving challenging, real-world problems - Genuine ownership of model improvement, performance and production engineering - See your work directly influence live commercial applications - Small, collaborative team where your contributions will have a visible impact If you're an ML Engineer who enjoys getting under the bonnet of complex models, understanding the maths behind them and making them perform better in production, click apply to learn more.