Sr. Machine Learning Engineer - Recommendations Platform at Twitter
San Francisco, CA, US

Recos Platform team builds recommendations platforms such as candidate generation and feature generation engines for product teams. The unrivaled challenges that we face at Twitter are both the data scale and the real-time nature of the product. How do you find the most meaningful content among hundreds of millions of new tweets for hundreds of millions of users every day at Twitter? We build large scale personalized recommendation engines utilizing different kinds of signals such as social network, user activity, and geolocation. Most of our work is about recommendation systems, machine learning, graph algorithms, distributed systems, and social graph analysis.

What You'll Do:

Apply your engineering skills to either improve existing recommendation systems, unlock new directions or provide entirely new ML solutions in recommendation systems within Twitter. You will work closely with live production systems and product teams, and deliver ML solutions at scale within the Twitter tech stack.

Who you are:

A machine learning software engineer with a passion for working on exciting algorithmic and deep infrastructure issues in ML environments.

    Thrive on working in concert with other smart people, including from distributed offices.
    Communicate fluidly, at the level of your audience, and seek to understand and be understood.
    Have the ability to take on complex problems, learn quickly, iterate, and persist towards a good solution.
    Take pride in polishing and supporting our products.
    Work hand-in-hand with modeling engineers and data scientists, and your passion is to enable them with better infrastructure.

Requirements:

BS, MS or PhD in Computer Science with 5+ years experience or equivalent experience.

    Fluent in one or more languages like Java, Scala, C++, Python
    Experience with Hadoop, Pig or other MapReduce-based architectures
    Knowledgeable of core CS concepts such as common data structures and algorithms
    Comfortable conducting design and code reviews