The company is building its own fintech platform from scratch, a large international initiative where engineers shape the architecture and build core fintech components designed to run at the scale of the company.
The role sets the ML architecture and modelling standards that credit decisioning will rely on, and the team wants these standards to be reused by others across the platform.
You will define modelling, validation, monitoring and responsible-AI and model-risk practices for credit-scoring and MCA merchant-underwriting models.
You will work with Credit Risk and the engineering teams that put models into decisioning services.
The stack includes the feature store, model registry and serving, MLflow and Databricks, agent orchestration graphs, eval harnesses, and model monitoring and tracing tools.
The company is building a lean engineering organisation in Poland with AI-native engineering at its core.
Responsibilities
Own the full model lifecycle: feature selection, training method, validation, champion/challenger testing, deployment and post-production monitoring for drift and stability.
Work with Data Engineering to define point-in-time-correct, production-grade features, and shape the feature-store roadmap around modelling needs.
Set documentation, model card and adverse-action reasoning standards so the models hold up to regulatory scrutiny.
Design agentic harnesses and loop or graph agent workflows tied to the feature store, covering model iteration (experiment, evaluate, retrain, promote) and model monitoring and observability.
Design agent loops and graphs around context, tools, state, evals and guardrails so they are reliable, observable and cost-controlled, and codify them as reference patterns for the function.
Lead the ML engineering team by setting technical direction, reviewing designs and code, and mentoring Senior and Mid ML Engineers.
Requirements
You enjoy building ML systems that perform in production and stand up to audit.
You design and build hands-on, using AI as a daily tool.
7+ years of machine learning engineering in production systems, including 3+ years in credit-scoring, risk modelling or decisioning ML on Spark/Databricks or a comparable lakehouse.
7+ years of Python and ML frameworks such as scikit-learn, XGBoost and PyTorch.
Strong credit-risk and underwriting model design skills, covering scorecards, gradient boosting and survival or behavioural models.
Solid model validation skills: back-testing, champion/challenger, population stability index (PSI) and calibration.
Experience with point-in-time-correct feature engineering, feature-store consumption and MLOps tooling such as MLflow and CI/CD for models.
Fluent in explainability and responsible-AI methods for regulated decisioning, including SHAP, model cards and adverse-action reasoning, plus working knowledge of model-risk frameworks.
3+ years with agentic coding tools and loop and graph engineering, including agentic harnesses, context and tool design, evals, guardrails and iteration, with a deep understanding of LLM behaviour and trade-offs.
Staff-level engineer who sets standards through influence rather than authority and mentors through code and design review.
English at B2 level or above.
Conditions
B2B contract.
Hybrid work, ideally one day a week in the Warsaw or Kraków office.
Equal opportunity applies to all candidates.
Benefits
Architectural authority from the start, with your design decisions expected to last for years.
Greenfield work with high stakes, a modern stack and space to try new technologies and bring them into production.
A collaborative, fast-moving team where autonomy and accountability go together.