Build and maintain machine learning models that predict the infrastructure cost impact of A/B experiments, translating experimentally observed signals (e.g., request volume changes) into business and system metrics (e.g. projected annualized costs)
Drive adoption of infrastructure metrics within the experimentation community through analysis, consultation with experiment owners, documentation, and training
Partner with platform teams (Observability, Experimentation Platform) to improve the quality and coverage of infrastructure usage data feeding our models
Extend our measurement framework to new metrics (e.g., latency) and new experiment types (e.g., infrastructure canary tests)
Champion an infrastructure lens within the broader experimentation community, helping shift culture toward reasoning about the full ROI and infrastructure impact of experiments
Connect with the larger analytics and experimentation communities at Netflix to bring visibility to our work and learn from others
Requirements
Experienced in experimentation methodology and causal inference, with a strong foundation in A/B testing, treatment effect estimation, and statistical significance
Experienced in building and maintaining machine learning models in production, including the full lifecycle of training, evaluation, monitoring, and continuous improvement
Fluent in Python and SQL, with experience engineering data pipelines and working with large-scale data systems
A strong collaborator who thrives in horizontal roles with broad stakeholder surfaces, comfortable influencing decisions through data and analysis rather than direct authority
An exceptional communicator who can flex between technical and non-technical audiences, translating statistical concepts for software engineers and business leaders alike
Comfortable with messy, incomplete data environments and able to balance short term execution with a drive to improve data quality over time
A strong product thinker who views data science outputs as products, taking an end-to-end ownership mindset from data quality through to user adoption
Comfortable with ambiguity, and thrive with minimal oversight and process
Curious about infrastructure systems; prior experience in the infrastructure domain is a strong plus but not required; the ability and motivation to learn is essential
Tech Stack
Python
SQL
Benefits
Health Plans
Mental Health support
401(k) Retirement Plan with employer match
Stock Option Program
Disability Programs
Health Savings and Flexible Spending Accounts
Family-forming benefits
Life and Serious Injury Benefits
Paid leave of absence programs
Full-time hourly employees accrue 35 days annually for paid time off