Machine Learning Engineer

Machine Learning Engineer (2–4 Years Experience)

London | Hybrid | £50,000 – £60,000 + stock options + Benefits

Company Overview

We’re an award-winning FinTech building fairer and smarter lending solutions. Our embedded and direct lending products rely on advanced machine learning models trained on billions of financial transactions.

As we scale, we are investing in strengthening our core ML capabilities — starting with a critical initiative: improving and scaling our transaction categorisation models, which underpin our credit risk, affordability, and behavioural analytics systems.

The Role

We are looking for a hands-on Machine Learning Engineer with 2–4 years of experience to join our growing ML team.

Your first major focus will be transaction categorisation: designing, training, and optimising classification models that accurately label and structure raw banking transaction data. This work directly impacts affordability assessments, credit models, and partner reporting.

You’ll work closely with credit analysts, data engineers, and the Lead ML Engineer to build scalable, explainable, and production-ready ML systems. This role is ideal for someone who enjoys applied machine learning, classification problems, and building robust systems in real-world environments.

What You'll Do

• Design and improve transaction categorisation models (multi-class classification problems).
• Work with structured financial transaction data (merchant strings, metadata, timestamps, behavioural features).
• Apply techniques such as:
• Gradient boosting (e.g., XGBoost)
• Tree-based models
• NLP approaches for merchant text classification
• Embedding techniques where appropriate
• Engineer and test new features to improve model accuracy and generalisation.
• Evaluate model performance using appropriate classification metrics (precision/recall, F1, confusion matrices).
• Ensure models are explainable and interpretable for risk and compliance teams.
• Collaborate with engineering for production deployment (deployment handled by engineering team).
• Contribute to ongoing improvement of credit and behavioural models beyond categorisation.

What We're Looking For

• 2–4 years of hands-on machine learning experience in production or applied environments.
• Direct experience working on classification or categorisation problems (ideally transaction, text, or behavioural categorisation).
• Strong proficiency in Python and ML libraries (e.g., XGBoost, scikit-learn, pandas).
• Comfortable working with SQL and large datasets (BigQuery experience a strong plus).
• Experience evaluating model performance rigorously and iterating based on data.
• Strong analytical thinking and attention to detail.
• Able to explain model outputs and decisions clearly to non-technical stakeholders
• Comfortable working in a small, fast-moving team.

Nice to Have

• Experience with transaction data, financial data, or open banking datasets.
• Familiarity with GCP, especially BigQuery and CloudRun.
• Exposure to NLP techniques for messy text classification.
• Experience building reusable model pipelines.
• Background in fintech, lending, or other regulated environments.

What We Offer

• Ownership of a high-impact ML initiative from day one.
• Opportunity to work on real-world credit systems used by live partners.
• A small, highly collaborative team where your work directly influences business outcomes.
• Competitive salary, stock options, and benefits.
• Hybrid working and flexibility.

Interview Process:

1. Recruiter Call – Background and fit discussion
2. Technical Interview – ML design and classification deep dive
3. Practical Task – Transaction categorisation modelling exercise
4. Final Interview – Collaboration, explainability, and business alignment