Machine Learning Engineer II

Affirm Affirm · Fintech · Canada, United States · Remote · Checkout

Machine Learning Engineer II at Affirm to build and improve AI systems for automating customer operations like disputes, returns, fraud, and chargebacks. Responsibilities include developing AI systems for dispute/chargeback handling, building models for automated refunds, creating evidence extraction pipelines using LLM-powered workflows, prototyping and deploying models, and collaborating with cross-functional teams. Requires 2+ years of ML engineering experience, strong Python, experience with tabular classification models, LLM APIs (OpenAI, Anthropic), prompt engineering, orchestration frameworks (LangChain, LangGraph), unstructured data processing, and ML lifecycle tooling.

What you'd actually do

  1. You will develop AI systems that automate dispute and chargeback handling using structured evidence and business logic, creating a better experience for our customers.
  2. You will build models that automate refunds, getting money back to our customers faster.
  3. You will build and maintain evidence extraction pipelines that process unstructured data using LLM-powered workflows to produce structured, actionable outputs.
  4. You will prototype new modeling ideas, run offline experiments, and drive the best-performing approaches into production with appropriate risk controls.
  5. You will collaborate across Engineering, Servicing Operations, Product, and ML Platform to define requirements, evaluate tradeoffs, and communicate results clearly to both technical and non-technical audiences.

Skills

Required

  • 2+ years of experience as a machine learning engineer
  • Strong Python skills
  • experience writing production-quality code
  • Experience building and evaluating models for tabular classification problems
  • Experience building applications with LLM APIs (e.g., OpenAI, Anthropic)
  • structured extraction
  • prompt engineering
  • orchestration frameworks like LangChain or LangGraph
  • Familiarity with document and unstructured data processing (PDF/image extraction, text parsing, or similar)
  • Experience with ML lifecycle tooling for training orchestration, experimentation, and model monitoring (e.g., Kubeflow, Airflow, MLflow, or equivalent internal platforms)
  • mastered taking a simple problem or business scenario into a solution that interacts with multiple software components, and executing on it by writing clear, easily understood, well tested and extensible code
  • comfortable navigating a large code base, debugging others' code, and providing feedback to other engineers through code reviews

Nice to have

  • gradient-boosted decision trees like LightGBM/XGBoost/CatBoost
  • AI-powered developer tools (e.g., Claude Code, Cursor, or similar)

What the JD emphasized

  • production-quality code
  • production
  • production

Other signals

  • LLM-powered workflows
  • automating customer operations
  • production ML systems