Staff Machine Learning Engineer

Skills

About the Role

You will design, build, deploy, and operate machine learning systems for fraud, scams, account takeovers, payment abuse, and other financial-risk use cases. You will own production ML systems, translate models into risk controls, develop LLM-based investigation agents, build evaluation frameworks, and ensure explainability, security, privacy, and reliability in regulated workflows.

Requirements

• Significant professional experience in machine learning engineering or applied data science
• Strong Python skills
• Experience with PyTorch, TensorFlow, XGBoost, LightGBM, or scikit-learn
• Knowledge of supervised learning, anomaly detection, representation learning, class-imbalanced modeling, calibration, and changing data distributions
• Experience using LLM coding tools and building AI-integrated workflows
• Familiarity with SHAP, feature attribution, reason-code generation, and model scorecards
• Experience deploying production LLM agents with tool calling and retrieval-augmented generation
• Experience integrating LLM agents with internal systems, APIs, databases, search tools, or decision engines
• Understanding of LLM-agent evaluation, reliability, observability, permissions, and failure handling
• Strong communication and collaboration skills

Responsibilities

• Design and deploy machine learning models for financial-risk use cases
• Own feature pipelines, training workflows, model serving, decision integrations, monitoring, alerting, drift detection, retraining, and incident response
• Translate models into production risk controls
• Work with risk operations to improve workflows, explainability, labels, and training data
• Apply AI-assisted development across implementation, testing, debugging, analysis, and documentation
• Develop AI-powered investigation and risk capabilities
• Take research-stage models into reliable production systems
• Ensure models and decision systems are explainable, traceable, and documented
• Design and deploy LLM-based agents for risk operations
• Build agent architectures with tool calling, retrieval-augmented generation, orchestration, memory, guardrails, and human review
• Build evaluation frameworks for LLM agents
• Implement permission controls, audit logs, privacy protections, prompt security, fallbacks, and escalation paths

Benefits

• Performance bonus
• Long-term incentives
• Medical benefits
• Financial benefits
• Other benefits