Machine Learning Engineer II, Measurement and Data Science

Job Overview

Come help build ML-based measurement at internet scale. This is a green grass problem without a known answer or a pattern to follow.

What You’ll Do

You will design, launch, own and evolve software that computes estimated impact of Amazon ads. Hundreds of thousands of advertisers will use your work every day to decide where to invest next. You'll work closely with our top-notch team of scientists and economists to invent, build and try ML-based approaches, and iterate on what works best. You'll help define not only how we compute the estimates, but how do we know we're right. These are really hard questions that are unique to Amazon, so there isn't a footprint you can copy. You'll build petabyte-scale measurement pipelines, as well as advance supporting services and frameworks. You'll get feedback from principal and senior engineers, as well as help junior engineers grow.

What We Do

We enable advertisers to optimize ad spend and allocate budgets effectively by providing accurate, actionable and timely conversion measurement for all Amazon ad products. We're in the early stages of work, where we invent and try a lot of new approaches. We use a combination of machine learning (ML)-based and deterministic techniques to produce the estimates that are fastest in the industry without compromising quality. We constantly invent on our cutting-edge event-driven architectures to stay ahead of growing scale.

Team Charter

The charter of this team is focused on computing estimated conversions. We own the math, the algorithms, and the design. We work with a variety of technologies, such as AWS EMR, Batch, Scala, Spark, and PyTorch. Science is invented in the same team.

What We Offer

We are a company of builders who bring varying backgrounds, ideas, and points of view to inventing on behalf of our customers. Our diverse perspectives come from many sources including gender, race, age, national origin, sexual orientation, culture, education, and professional and life experience. We are committed to diversity and inclusion and always look for ways to scale our impact as we grow.

Key Job Responsibilities

1. Design, develop, launch, and maintain measurement applications that compute accurate estimates of advertising impact.
2. Work with the data to understand advertiser needs and develop the most efficient ways of handling it.
3. Closely partner with the science team on all aspects, from featurization to training to estimating to knowing we're right.
4. Write code in languages like Scala, Java, or Python.
5. Support and improve what you built.

Minimum Requirements

• 3+ years of non-internship professional software development experience.
• 2+ years of non-internship design or architecture (design patterns, reliability, and scaling) of new and existing systems experience.
• Experience programming with at least one software programming language.
• 3+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience.
• Bachelor's degree in computer science or equivalent.

Amazon is committed to a diverse and inclusive workplace. Amazon is an equal opportunity employer and does not discriminate on the basis of race, national origin, gender, gender identity, sexual orientation, disability, age, or other legally protected status. If you would like to request an accommodation, please notify your Recruiter.