Machine Learning Engineer, Financial Crime

It all started with an idea in 2021. Initially built to democratize Bitcoin in France, Bitstack has gone from a simple product that invests spare change in bitcoin, to a crypto-native bank bringing a better way to save, spend, and grow money to our 300K+ users. We want to rebuild finance from the ground up to empower people to achieve financial freedom by participating in a more open and inclusive financial system. We're a small, ambitious team of operators. We believe in moving fast, sweating the details, and creating products people truly love. At Bitstack, you'll have the opportunity to shape not just our products, but the culture and practices of our entire organization. Bitstack is backed by YC, leading VCs like 13books Capital, Stillmark, Serena, and many more. We are building a product our users love and we are just getting started

We're hiring a Machine Learning Engineer focused on Fraud & Financial Crime to build the models and systems that protect our users and secure Bitstack's financial ecosystem. You'll develop models, build real-time monitoring systems, and work closely with Product, Engineering, and Compliance to protect our users while enabling a seamless experience.

This is a high-impact role where your work directly contributes to the safety, integrity, and scalability of Bitstack's financial ecosystem.

• Build and refine fraud detection models (e.g., anomaly detection, behavioral scoring, supervised ML).
• Design real-time risk evaluation pipelines to flag suspicious activity before it impacts users.
• Partner with Engineering and Compliance to map fraud patterns, investigate cases, and translate insights into product or policy changes.
• Create data tools and dashboards to monitor fraud trends, model performance, and operational KPIs.
• Own the end-to-end lifecycle of fraud models: scoping, experimentation, evaluation, deployment, and monitoring.
• Conduct deep dives into user behaviour and transactional data to uncover vulnerabilities or emerging fraud vectors.
• Collaborate with internal teams to bring a data-driven approach to fraud prevention across the organisation.
• Own all fraud-related tooling and services at Bitstack (from model training to real-time scoring infrastructure).

You don't need to know everything - but experience with several of the tools below is strongly valued:

• Languages: Python (strong), SQL
• Frameworks / Libraries: scikit-learn, XGBoost/LightGBM, PyTorch/TensorFlow (nice to have)
• Data: Vector databases (Pinecone, Weaviate) and modern warehouses (Snowflake, BigQuery)
• Model Serving / Pipelines: FastAPI, Airflow, feature stores (e.g., Feast)
• Monitoring & Analytics: Metabase, Looker, or internal dashboards
• Bonus: experience with fraud tools, graph analysis, or real-time scoring systems

• Experience building and deploying ML models end-to-end, ideally in fraud, risk, fintech, or trust & safety.
• We're looking for someone who can operate with autonomy and own fraud ML systems end to end.
• Ability to deliver impact cross-functionally, across ML
• Strong expertise in Python and SQL, with the ability to turn messy data into clean, usable signals.
• Ability to explore large datasets, find patterns, and turn findings into actionable insights.
• Comfortable working with stakeholders from Engineering, Product, and Compliance.
• A strong sense of ownership - you enjoy operating with autonomy and driving projects from idea to production.
• Excellent communication skills and the ability to simplify complex data topics.
• Fluent English (French is a bonus)

• Introductory Call (45 min)
  
  An initial conversation with our CTO to get to know you better and share more details about the role, the team, and how we work at Bitstack
• System Design (60 min)
  
  A discussion focused on how you approach problem-solving, communicate your ideas, and prioritize decisions. This helps us assess your technical depth and ability to design scalable systems
• Take Home Task & Task Review Call (60 min)
  
  You'll receive a practical task to complete in your own time. Once submitted, we'll schedule a review call to walk through your solution together and explore your thought process in more detail
• Behavioural Interview (60 min)
  
  A conversation centered on your experiences, decision-making, and collaboration style. We'll discuss real-world scenarios to understand how you navigate technical challenges, teamwork, and stakeholder interactions.

Work from our office in Paris (Bourse) or Remote

Flexible working hours

One of the best health insurance plans with 100% employer contributions

Monthly team activities and bi-annual offsites

Swile meal vouchers

Competitive salary based on experience & Bitstack stock options

We hire talented and passionate people from all kinds of backgrounds, we don't care about your age or where you studied. If you're excited about the role but your past experience doesn't match every bullet point in the job description, we still encourage you to apply. What matters most is that you're a builder at heart, you share our values, and you're enthusiastic about our mission.