Staff ML Platform Engineer: AI-Native FinTech, Reg-Ready

InPost; Warszawa, Kraków; Hybrid; PLN 260,000 - 360,000; Full time

Job summary

InPost Pay is building its own fintech platform from the ground up in Poland. The mission is to set ML architecture and modelling standards for credit decisioning, and to own the end-to-end model lifecycle with a strong focus on feature stores, monitoring, and responsible-AI practices.

You will lead an engineering team, mentor senior engineers, and partner with data and risk teams to deploy scalable, auditable models in a hybrid work environment.

Qualifications

• Experience building ML systems that endure production audits.
• We design and build with AI as a daily tool.
• 7+ years in ML in production, 3+ years in credit-risk modelling or decisioning.
• Expert in Python and ML frameworks (scikit-learn, XGBoost, PyTorch).
• Strong in credit-risk and underwriting model design.
• Model validation methodology knowledge: back-testing, PSI, calibration.
• Feature-store concepts and MLOps tooling (MLflow, CI/CD for models).
• Explainability and responsible-AI for regulated decisioning.
• Agentic coding tools and loop/graph engineering; deep LLM knowledge.

Responsibilities

• Set ML architecture and modelling standards for credit decisioning.
• Own end-to-end model lifecycle from feature selection to deployment and monitoring.
• Partner with Data Eng to define production-grade features and feature-store roadmap.
• Establish documentation, model cards and adverse-action reasoning standards.
• Architect agentic harnesses and loop/graph workflows.

Skills

Production ML systems

Hands-on implementation

Mentoring engineers

Staff-level influence

Regulatory model validation

Explainability

LLM engineering

Agentic engineering

FinTech / payments ML

Tools

Python

MLflow

Databricks

Spark

Scikit-learn

XGBoost

PyTorch

Job description

InPost Pay is building its own fintech platform from the ground up in Poland. The mission is to set ML architecture and modelling standards for credit decisioning, and to own the end-to-end model lifecycle with a strong focus on feature stores, monitoring, and responsible-AI practices.

You will lead an engineering team, mentor senior engineers, and partner with data and risk teams to deploy scalable, auditable models in a hybrid work environment.