Applied Scientist II — AI-Driven Ad Campaign Optimizer

Amazon; New York (NY); On-site; USD 172,000 - 223,000; Full time

Job summary

Amazon Ads is seeking an Applied Scientist II to build the AI behind Auto Optimization: agentic systems that automatically optimize advertisers' campaigns. These agents observe performance, reason about changes, and act continuously as market conditions shift.

Optimizing campaigns requires coordinating targeting, bids, budgets, and placements across campaigns, with adaptivity to different advertiser goals. You will ship live features that impact millions of campaigns.

Qualifications

• 3+ years of building models for business application experience.
• PhD, or Master’s degree and 4+ years of CS, CE, ML or related field experience.
• Experience programming in Java, C++, Python or related language.
• Experience in algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, high-performance computing.

Responsibilities

• Build agentic systems that automatically optimize campaigns across targeting, bids, budgets, and placements.
• Encode the dynamics of the auction and marketplace to balance advertiser return, shopper experience, and marketplace health.
• Turn raw signal into intelligence by defining and curating datasets that train and evaluate these agents.
• Push the frontier of agent design with core planning, tool use, and long-horizon reasoning, delivering production-quality code.
• Set the bar for trust by developing evaluation and safety methods for live campaigns.
• Grow with the team by contributing to our scientific agenda and sharing work with the broader community.

Skills

Business-modeling experience

Coding: Java C++ Python

Algorithms & data structures

Agentic/LLM-based systems

Education

PhD or Master’s in CS/CE/ML or related field

Tools

Unix/Linux

Job description

Amazon Ads is seeking an Applied Scientist II to build the AI behind Auto Optimization: agentic systems that automatically optimize advertisers' campaigns. These agents observe performance, reason about changes, and act continuously as market conditions shift.

Optimizing campaigns requires coordinating targeting, bids, budgets, and placements across campaigns, with adaptivity to different advertiser goals. You will ship live features that impact millions of campaigns.