Data Scientist, Production Analytics

Carvana•Tempe, AZ

About The Position

At Carvana, we’re changing the way people buy and sell cars. With an ambitious vision and a fundamentally different approach designed to be fun, fast, and fair, Carvana became the fastest-growing automotive retailer in history. We expanded nationally, went public on the New York Stock Exchange, sold our 1 millionth car, and reached the Fortune 500, all in just eight years. Today, with 4 million retail customers and counting, Carvana is both the fastest-growing and the most profitable public automotive retailer, and we’re just getting started. We continue to raise the bar for our customers as we tackle the enormous opportunity still ahead in the largest consumer vertical. Working here means being part of a team that embraces change, celebrates creative problem solving, and always strives to be better. At Carvana, you’ll have the opportunity to take on meaningful challenges, learn quickly, and help shape the future of automotive retail. If you’re driven to grow and make an impact as part of a collaborative team, you’ll fit right in. Learn more about what it’s like to work here from the people that already do . About the team and position Have you ever wished the analytics projects you worked on could have visible and tangible benefits? The production analytics team owns advanced analytics (data science, optimization, data mining, etc.) in Carvana’s vertically integrated vehicle inspection and reconditioning centers. Our team can walk through these centers and see our work in action. We build predictive models that support decision-making across our reconditioning operations. These models leverage vehicle characteristics and operational data to generate insights that drive business outcomes. The work can be high-visibility and high-impact. Our models directly influence operational strategy and performance measurement. As a Data Scientist, you'll work closely with our Associate Director to build, maintain, and improve these production models. You'll develop both new models and maintain existing ones, with a focus on neural network architectures that power our production systems. This is a hands-on role where you'll see your work deployed and making a difference.

Requirements

Bachelor's degree in Computer Science, Statistics, Mathematics, Data Science, or a related quantitative field.
Strong understanding of neural networks. You should know how they work in practice, not just theory (architecture design, training dynamics, hyperparameter tuning, regularization).
Proficiency in Python for data science (pandas, numpy, scikit-learn, TensorFlow/Keras or PyTorch).
SQL skills for data exploration and validation (Snowflake experience preferred).
Strong statistical foundations (hypothesis testing, regression, experimental design).
Experience with Git-based version control, branching, and pull requests.
Skeptical mindset about data quality—you don't assume data is correct, you validate it.
Excellent communication skills to explain technical concepts to non-technical audiences.
Eagerness to learn and grow through mentorship and code reviews.
A relentless drive to push work into production and see tangible impact.

Nice To Haves

Hands-on experience with TensorFlow/Keras (preferred) or PyTorch for building production models.
Experience with LightGBM or XGBoost for gradient boosting models.
Familiarity with Snowflake or other cloud-based data warehouses (Google Cloud Platform, AWS, Azure).
Understanding of model deployment and MLOps practices.
Experience with hyperparameter optimization frameworks (e.g., Hyperopt, Optuna).
Knowledge of data validation and testing techniques.
Exposure to manufacturing, operations, or automotive analytics environments.
Experience building models for cost prediction or resource optimization.
Understanding of Agile/iterative development practices.
Familiarity with containerized code (Docker, Kubernetes).

Responsibilities

Building and maintaining predictive models: Developing neural network models (primarily TensorFlow/Keras) that leverage vehicle characteristics and operational data
Building models that support operational decision-making and performance insights
Maintaining and improving existing production models as business needs evolve
Using LightGBM and linear/logistic regression for benchmarking and explanatory analysis
Implementing hyperparameter optimization and model tuning strategies
Monitoring model performance and identify when recalibration is needed
Performing rigorous exploratory data analysis: Investigating data quality before training models—validating that data matches operational reality
Grounding expectations to what's actually happening in the business, not just what the data says
Identifying anomalies, edge cases, and data quality issues that could impact model performance
Creating visualizations that communicate patterns and insights to stakeholders
Documenting data validation findings and decisions
Deploying and monitoring production models: Working with engineering teams to integrate models into operational systems
Tracking prediction accuracy and model drift over time
Developing validation frameworks to ensure models perform as expected
Iterating on models based on production performance and changing business needs
Collaborating and communicating: Translating business problems into modeling approaches
Explaining model predictions and limitations to non-technical stakeholders
Partnering with operations teams to understand domain context
Documenting model architecture, assumptions, and performance characteristics
Working under the technical direction of our Associate Director with regular guidance and code reviews