I’m upgrading our risk-and-compliance platform and need an EU-based engineer who can blend robust software craftsmanship with hands-on machine-learning expertise. The mandate is to automate decision-making across key customer-facing features by fusing classical ML techniques with Large Language Models. You’ll design the end-to-end pipeline in Python, harnessing PyTorch, TensorFlow or scikit-learn to train, evaluate and continuously improve models, then ship them as production-ready services that scale and stay maintainable. Solid experience integrating models into distributed systems, setting up meaningful evaluation metrics and communicating findings to non-technical stakeholders will be crucial, as you’ll work side-by-side with product, legal and data teams. Deliverables I’ll review: • Clean, well-documented Python codebase that builds, tests and deploys the chosen models • Containerised service or microservice exposing the decision-automation endpoint • Evaluation report covering model performance, drift-monitoring strategy and rollback plan If you have 4+ years of professional software engineering in production ML settings and can articulate your design choices clearly in English, I’d like to hear how you would tackle this challenge.