Overview In this role you will own the science behind the Intelligence Flywheel that powers next-gen smart ads. You'll turn static logic into self-improving, autonomous decision-making, enabling trusted advertiser experiences. You will build predictive models, design intervention policies, and establish causal attribution for autonomous optimization. You'll integrate cross-model signals and define reward schemas, deploying your work in production alongside software engineers. This is a high-impact, autonomous role shaping AI-powered ad products at scale.
Compensation / Benefits- health insurance: medical, dental, vision, prescription
- 401(k) matching
- paid time off
- parential leave
- RSUs
- sign-on payments
Responsibilities- Build predictive issue-detection models to flag under-delivery, over-delivery, and performance degradation before they affect advertisers
- Design the intervention-selection policy as a contextual bandit to guide autonomous actions
- Establish causal attribution to separate intervention effects from organic performance changes
- Integrate cross-model signals to form a unified advertiser-intelligence view and adapt models via fine-tuning or thin adaptation
- Close feedback loops by defining reward schemas and linking creative quality to campaign outcomes
- Collaborate with software engineers to ensure models are deployed and served in production
- Set the science roadmap for the Model Layer and guide multi-engineering workstreams
- Work on ambiguous business problems and translate them into scalable AI-powered solutions
Key requirements- 3+ years of building ML models for business applications
- PhD, or Master's degree and 6+ years of applied research experience
- Experience programming in Java, C++, Python or related language
- Experience with neural deep learning methods and machine learning
- analytical thinking in ambiguous problems
- strong collaboration with cross-functional teams
- ability to work autonomously and deliver impact
- TensorFlow
- scikit-learn
- Spark MLLib
