Role overview
At Modulai, we focus 100% on solving problems with machine learning (ML). We work in teams on a project basis, for clients, as part of the core team in startups where we have long-term engagements, and we also build our own ML products.
Learning and teamwork are central to how we work. Everyone in the team is or will soon be a full-stack ML engineer capable of scoping and developing end-to-end ML solutions. You should be able to do end-to-end machine learning products by yourself but never do it because we always work in teams. If there is data, we will do ML on it!
What you'll work on
- Design, train, evaluate, and deploy ML models for transaction-level fraud detection (primarily tabular data).
- Analyze large-scale transaction datasets to identify patterns, leakage, bias, and data quality issues.
- Build and maintain production ML services (real-time and batch).
- Implement robust ML pipelines, model monitoring, and experiment frameworks.
- Collaborate directly with client engineers, data scientists, and risk teams.
- Translate complex technical concepts and results into clear, actionable insights for technical and non-technical stakeholders.
- Operate within strict requirements for reliability, explainability, traceability, and compliance.
What we're looking for
- Production-grade Python and solid ML fundamentals (XGBoost/LightGBM, Scikit-learn, feature engineering, imbalanced datasets)
- Experience building and shipping ML-powered APIs (FastAPI/Flask), Docker, CI/CD, and distributed data processing (PySpark/SQL)
- Strong stats foundation: experimental design, bias/leakage detection, time-dependent validation
- Hands-on MLOps experience — feature stores, Airflow/Kubeflow, model monitoring, real-time inference, A/B testing
- MSc or Ph.D. in a quantitative field
- Excellent understanding of a broad set of ML algorithms and frameworks
- A passion for lean, clean, and maintainable code
- The desire to grow and to share insights with others
Domain experience: Fraud detection, payments, fintech, or credit risk. You've worked with cost-sensitive decisions, highly imbalanced data, and models that directly impact business risk.
How you work: You communicate clearly with engineers, product, and compliance stakeholders alike. You write good documentation and can hold your own in architecture discussions.