Machine Learning Engineer in London

As an ML Engineer at HASH you will develop and iterate on models for classification, prediction, recommendation, ranking, anomaly detection, optimization and more. You will work closely with product, engineering, and customers to define problems, explore data, prototype solutions, and measure impact. We are hiring for this primarily remote role across both Germany and the UK (existing right-to-work required). Successful candidates are also welcome to work from our Berlin office, should they wish.

Requirements

• Have 3+ years of experience in a Data Scientist / Applied Scientist / ML Engineer role.
• Are comfortable framing ambiguous problems and pushing for clarity on goals and constraints.
• Are fluent in Python and the standard data stack (e.g., pandas, NumPy, scikit-learn, Jupyter; plus at least one of PyTorch/TF/JAX, etc.).
• Are comfortable working with SQL (or similar) to pull and shape data.
• Care about clarity and communication: you can explain trade-offs, caveats, and uncertainty to non-specialists.
• Think pragmatically: you know when to ship a simple model and when it's time to reach for something more advanced.
• Hands-on experience in supervised learning (classification/regression), including feature engineering and regularization.
• At least one of: time series, recommender systems, or ranking/optimization problems.
• Model evaluation, validation, and experiment design (A/B testing, cross-validation, backtesting).

Nice-to-have

• Vector search, embeddings, or RAG-style systems.
• Causal inference and robust experimentation in messy environments.
• Optimization / operations research style problems.
• Building data products or AI features inside SaaS or platform products.
• B2B / enterprise environments with complex domains and heterogeneous data.
• Knowledge graphs or graph-based modeling.
• Evaluating and monitoring LLM- or agent-based systems.

What you will work on

• Work with stakeholders to translate product and business goals into clear modeling objectives and success metrics.
• Explore and evaluate available data sources (internal and external), identifying gaps and opportunities.
• Choose appropriate modeling approaches (simple baselines → advanced methods) and keep complexity justified.
• Build, iterate on, and validate models for classification and scoring, prediction and time-series forecasting, recommendation and ranking, anomaly detection and segmentation.
• Collaborate closely with MLOps.
• Package models and pipelines so they can be handed off cleanly to MLOps for deployment.
• Define clear contracts: inputs/outputs, service-level expectations, monitoring signals, and retraining triggers.
• Document assumptions, data expectations, and model behavior in a way that’s usable by others.
• Own evaluation and experimentation.
• Design and run experiments (A/B tests, offline evaluations, backtests) to understand model impact.
• Build evaluation suites and dashboards to track model performance over time (quality, fairness, stability, drift).

Contribute to HASH's AI product

• Work with the product and engineering teams to make HASH's platform better for data scientists: feature engineering workflows, evaluation tooling, data access patterns, etc.
• Help define best practices for responsible, governance-first model development: reproducibility, provenance, and explainability.

We offer leading equity-weighted total compensation, including competitive salaries and tax-advantaged options. We also provide employer pension contributions, at least 30 days paid time off per year, and twice-yearly in-person team retreats around the world.

HASH provides an open-source platform which helps firms integrate both structured and unstructured information into knowledge graphs that support simulating, optimizing and automating processes. Our mission is to solve information failure, and help everybody make the right decisions. To that end, we are unapologetically excited. Actions speak louder than words, and we measure performance by output. We prioritise speed, and measure product delivery timelines in hours and days, not months and years. We value high-energy, high-expectations people who do what they say and say what they mean. We are committed to building a high-commitment, high-trust environment, and believe that the best teams are most productive together, in-person.