Pinterest is looking for a Machine Learning Engineer to build out our simulation and AI capabilities. You'll design and implement systems that model the CTV advertising ecosystem — auction dynamics, bidding strategies, campaign outcomes, and counterfactual scenarios — and develop AI-driven tools that accelerate how we build, test, and deploy ML systems.
What You'll Do
- Design and build simulation environments that model CTV auction mechanics, inventory supply, and advertiser competition
- Develop counterfactual and what-if frameworks for evaluating bidding strategies, budget allocation, and pacing algorithms offline
- Build AI agents that explore strategy spaces, generate hypotheses, and automate experimentation within simulated environments
- Use LLMs and generative AI to accelerate internal ML workflows — synthetic data generation, code generation, automated analysis, and rapid prototyping
- Use simulation to de-risk ML model deployments — validate new bidding and optimization strategies before they touch live traffic
- Define the technical direction for simulation and AI infrastructure and mentor engineers on the team
What We're Looking For
- Strong production Python skills and experience building simulation or modeling systems
- Deep understanding of probabilistic modeling, stochastic processes, or agent-based simulation
- Hands-on experience with modern AI tools: LLMs, code generation, agentic workflows — and good judgment about when they help vs. when they don't
- Adtech experience: you understand auction theory, RTB mechanics, and the dynamics of programmatic advertising
- Ability to translate business questions into rigorous simulation frameworks
- Clear written communication: you'll be defining new technical directions and need to bring others along
- Ownership: you scope, design, and ship systems end-to-end with minimal direction
Nice to Have
- Causal inference — uplift modeling, synthetic controls, difference-in-differences, or incrementality testing
- Experience with discrete event simulation, Monte Carlo methods, or digital twins
- Reinforcement learning — using simulated environments for policy learning and evaluation
- Experience building agentic AI systems or multi-agent simulations
- Big data experience with Scala and Spark
- Systems programming experience in Zig or similar (C, C++, Rust)
- MLOps experience — model deployment, monitoring, and pipeline orchestration on AWS
Technical Stack
- Python
- LLMs
- AWS
- Scala
- Spark
- Zig
- C
- C++
- Rust
Benefits & Compensation
- Compensation: $123,696 — $254,667 USD + equity: This position is eligible for equity
Work Mode
This role is remote.
Pinterest is an equal opportunity employer and makes employment decisions on the basis of merit. All qualified applicants will receive consideration for employment without regard to race, color, ancestry, national origin, religion or religious creed, sex (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender, gender identity, gender expression, age, marital status, status as a protected veteran, physical or mental disability, medical condition, genetic information or characteristics (or those of a family member) or any other consideration made unlawful by applicable federal, state or local laws.
