OKX - Senior Machine Learning Engineer, Compliance

• Design, build, and deploy ML models for compliance use cases including AML transaction monitoring, customer risk rating, KYC/KYB risk scoring, sanctions exposure detection, and SAR analytics, working closely with data scientists on model architecture and with data engineers on pipeline design. • Own the production infrastructure for compliance ML: feature pipelines, model serving, monitoring, drift detection, and retraining workflows. Models in a compliance context need to be reliable, auditable, and well documented, and you will be responsible for making sure they are. • Build and maintain internal ML tooling that the broader team depends on: reusable pipeline components, experiment tracking, model registries, and evaluation frameworks that raise the quality and speed of model development across the team. • Work with data scientists to take research-stage models into production, reviewing feature logic, validating pipeline assumptions, and bridging the gap between a notebook and a deployment that a compliance team can rely on. • Collaborate with compliance and legal stakeholders to ensure models are explainable and documented to the standard required for internal governance and regulatory review. • Keep a close eye on developments in compliance-relevant ML: graph neural networks for network-based AML detection, anomaly detection approaches for novel typologies, and emerging LLM applications in regulated environments, bringing relevant ideas into the team's work where they hold up to scrutiny. • What We Look For In You • 8+ years in ML engineering, data science, or a closely related field, with a strong track record of taking models from prototype to production in environments where reliability and auditability matter. We welcome candidates across seniority levels; scope will be calibrated to your experience. • Solid Python and experience with ML frameworks alongside hands-on MLOps practice including model deployment, monitoring, and CI/CD pipelines for ML workflows. • Experience with big data platforms such as Spark, Databricks, Hadoop, or MaxCompute, and comfort designing and optimising both batch and real-time data pipelines that feed production ML systems. • Demonstrable fluency with AI-assisted development: you use LLM coding tools before, you have built LLM-integrated pipelines or automation workflows in a professional context, and you have a clear, practical view on where these tools genuinely improve engineering output and where they introduce risk. • A good working knowledge of ML fundamentals across supervised, unsupervised, and anomaly detection methods, and an understanding of how different approaches translate to compliance problem formulations. • Familiarity with explainability frameworks such as SHAP or LIME, and an appreciation for what model governance and auditability require in a regulated environment. • Good communication skills and a collaborative working style, with the ability to work effectively alongside data scientists, compliance domain experts, and engineers who each bring a different perspective to the same problem. • Experience in financial services, fintech, or a crypto exchange, particularly in AML, KYC/KYB, transaction monitoring, or a related compliance domain, is a meaningful advantage. • Familiarity with the crypto ecosystem, on-chain data, blockchain analytics, or VASP regulatory frameworks is a plus and will give you a head start in understanding the data you will be working with.