Machine Learning Infrastructure Engineer - Ads Prediction (Revenue Science)
Who we are:
In Twitter, we serve billions of ad impressions and generate millions of dollars in revenue per day. Behind every ad, our ads prediction system, which is one of the largest ML services at Twitter, evaluates at least thousands of candidates behind the scene. Our mission is to make Twitter ads as relevant in the moment as organic content on Twitter.
What you will do:
In order to realize our mission, we need your help to design and build state-of-the-art large-scale machine learning systems and infrastructure to predict users’ engagement/conversion on different ads as accurately as possible. Although you will work on cutting-edge problems, this position is not a research position. You will participate in the engineering life-cycle at Twitter, including writing production code, conducting code reviews and working with our ML Modelers and alongside other infrastructure teams. You will be equally comfortable doing incremental quality work and building brand new systems to enable future quality improvements.
- Work with large unstructured and structured data sets (multi-terabyte+, 100MM+ daily transaction volumes)
- Design and evaluate novel approaches for handling high-volume real-time data streams in a machine learning environment
- Code using primarily Java and Scala; Map-Reduce frameworks such as Pig and Scalding; and scripting languages such as Python
- Learn new machine learning, data mining, and/or natural language processing techniques for a variety of modeling and relevance problems involving users, their tweets, their interests, twitter ads, relationship among entities
- Conduct online A/B testing, interpret and understand algorithm performance
Who you are:
You’re a back-end software engineer who wants to work on exciting algorithmic and deep infrastructure issues in ML environments.
- 3+ years relevant experience and MS or PhD in computer science
- Fluent in one or more object oriented languages like Java, Scala, C#, C++
- 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
- Experienced in operating Linux-based systems
- Passionate about working with large unstructured and structured data sets
San Francisco applicants: Pursuant to the San Francisco Fair Chance Ordinance, we will consider for employment qualified applicants with arrest and conviction records.
Engineering Hiring Process
Once your application is received, a recruiter will reach out pending your qualifications are a match for the role.
If your background is a match, you may have 1-2 technical phone interviews or be given the chance to provide a work sample depending on the role.
If the phone interviews go well or your work sample is strong, the final step includes interviews with 5-6 people held onsite in our office.
We're the People Team at Twitter. We Tweet about who we're hiring, what we're doing, and why you should work at Twitter! #LoveWhereYouWork
We're your one stop shop for anything University related. That means campus outreach, student advice/tips, & of course, our University Recruiting efforts!