Staff Data Scientist

About the Role

The Staff Data Scientist will play a key role in advancing IAA’s Machine Learning (ML) and advanced analytics initiatives. This role involves building supervised and unsupervised ML models, as well as developing AI agentic applications, with a strong business orientation and a track record of delivering data-driven services. The Staff Data Scientist will implement scalable, machine learning-based solutions to drive growth and operational efficiency across the organization.

Working closely with cross-functional teams within the IAA ecosystem, the Staff Data Scientist will help integrate these solutions into enterprise-scale services. This role requires deep knowledge of data best practices to ensure the most appropriate data sources are leveraged to answer business questions and perform advanced analysis of customer behavior, vehicle listings, and other business data.

Responsibilities

• Lead end-to-end data science projects, from problem framing and data exploration through model development, deployment, and monitoring

• Design and build supervised, unsupervised, and deep learning models to solve high-impact business problems

• Develop AI agentic applications and LLM-powered solutions to automate workflows and unlock new capabilities

• Design and analyze experiments (A/B tests, causal inference) to measure model and product impact

• Perform feature engineering, data validation, and quality assurance across large, complex datasets

• Partner with data engineering and ML platform teams to productionize models and ensure scalability, reliability, and monitoring

• Translate ambiguous business questions into well-scoped data science problems and communicate findings to technical and non-technical stakeholders

• Mentor junior data scientists and analysts, and contribute to best practices across the data science team

Required Qualifications

• 5–7 years of experience building and deploying production machine learning models, including supervised, unsupervised, and deep learning approaches

• Advanced proficiency in Python and SQL, with deep experience in ML libraries such as scikit-learn, PyTorch or TensorFlow, pandas, and NumPy

• Hands-on experience building AI agentic applications using LangChain, LangGraph, or similar frameworks, including integration with LLMs and external tools/APIs

• Hands-on experience with the full ML lifecycle: feature engineering, model training, hyperparameter tuning, evaluation, deployment, and monitoring

• Strong foundation in statistics, experimental design (A/B testing), and core ML algorithms (gradient boosting, neural networks, clustering, dimensionality reduction)