Senior ML Engineer – Real-Time Ad Tech & AI (Equity)

Pinterest; San Francisco (CA); On-site; USD 156,000 - 320,000; Full time

Job summary

tvScientific seeks a Sr. Machine Learning Engineer to build the ML systems behind its Connected TV ad-buying platform: real-time bidding, campaign optimization, and incrementality measurement at scale.

You will train, deploy, and monitor models deciding ads to show, when, and at what price, handling millions of bid decisions per second. Collaboration across a distributed team is essential.

Qualifications

• Proficiency writing production-grade Python for real-time systems.
• Strong understanding of experiment design and model evaluation.
• Experience with adtech or CTV is desirable.
• Ability to own problems end-to-end in a fast-moving environment.

Responsibilities

• Write production Python that powers real-time bidding, model training, and campaign optimization.
• Train, deploy, and monitor ML models that decide which ads to show, when, and at what price.
• Build and improve incrementality measurement systems to show true causal lift of spend.
• Design ML products across the ad-buying lifecycle: targeting, bidding, pacing, attribution.
• Use LLMs to build internal tools that accelerate ML development and testing.

Skills

Python production code

Statistics & ML fundamentals

AI tools awareness

Adtech/CTV experience

Clear written communication

Ambiguity tolerance

Bachelor's degree in CS/Math/Eng

4+ years experience

LLMs / AI coding assistants

MLOps / deployment

Education

Bachelor's degree in Computer Science, Mathematics, Engineering or related

Tools

Cursor / Copilot / Codex

SQL / Data exploration

Big data tools (Scala, Spark)

Job description

tvScientific seeks a Sr. Machine Learning Engineer to build the ML systems behind its Connected TV ad-buying platform: real-time bidding, campaign optimization, and incrementality measurement at scale.

You will train, deploy, and monitor models deciding ads to show, when, and at what price, handling millions of bid decisions per second. Collaboration across a distributed team is essential.