Staff Machine Learning Researcher, Cortex Recommender Systems

Company Description

Scale out Twitter’s recommender systems to hundreds of millions of users and trillions of records!

Twitter is what’s happening and what people are talking about right now. For us, life's not about a job, it's about purpose. We believe real change starts with conversation. Here, your voice matters. Come as you are and together we'll do what's right (not what's easy) to serve the public conversation.

The Position

Who we are:

We are a community of Machine Learning Researchers and Engineers, working to help Twitter leverage ML through a range of systems such as recommendations, safety, abuse, content understanding, ads and more. We operate at scale whilst ensuring fair and ethical use of our models and data.

We work collaboratively, often embedding among product teams, looking to apply the expertise of the individuals to improve our products and unlock new capabilities. We encourage publishing papers, but they are not the end goal, rather a by-product of us doing interesting work - the aim is to make a real-world impact!

The Recommender Systems Research Team, part of Cortex Applied Research, both performs fundamental recommender systems research and builds some of our most sophisticated recommender systems for the most important products at Twitter. We strive to understand the social network via state of the art ML and a deep understanding of our domain, and to recommend content that sparks joy. In this team, you'll be working closely with leading researchers and will have the opportunity to learn about the current state-of-the-art in ML, contributing to ambitious research projects based on the latest insights.

What you will do:

Apply your research and engineering skills to bring ML research ideas to production. You will work closely with your fellow researchers and engineers to scale up models for both training and inference on state-of-the-art hardware. You will contribute to the design of new systems and infrastructure to shape how ML is used across the business. Apply your wider experiences to provide perspective, and enable Twitter to benefit more rapidly from fundamental ML research. In addition, you will be directly contributing to research projects, strategic decisions and future roadmaps for products and technologies at Twitter. You will work on the handful of most important models at Twitter.

Who you are:

You are excited about the challenges of deploying the latest ML technologies at scale and enabling research to go from idea to practical impact. You have a deep understanding of the interplay between ML techniques and the hardware compute constraints. You’re a diligent listener, have the ability to communicate technical issues to non-experts, and can persuade others to follow your recommendations. You have a solid understanding of core ideas in machine learning and software engineering, keep up-to-date with the latest developments in the field and look for ways to apply them in the real world. You are deeply pragmatic, and know when to build for the long-term, and when to do quick experimentation to bring clarity in the short-term. 

Qualifications

Required Skills:

  • Masters’ or PhD in a Computer Science or Machine Learning related degree; or equivalent work experience
  • Experience optimizing models for training throughput or for inference.
  • Experience running benchmarks for model latency and throughput.
  • Experience building models in PyTorch or TensorFlow.
  • Practitioner of modern software development practices - e.g. TDD, CI, source control

Nice to haves:

  • Experience with Java or C++.
  • Experience with CUDA.
  • Experience optimizing models for inference through distillation, pruning, quantization and related techniques.
  • Good theoretical grounding in core machine learning concepts and techniques.
  • Publications in premier conferences such as NeurIPS, ICLR, ICML, CVPR, KDD, AAAI, IJCAI, etc., or on arXiv
  • Ideally, 4-6+ years work experience

Additional Information

We are committed to an inclusive and diverse Twitter. Twitter is an equal opportunity employer. We do not discriminate based on race, ethnicity, color, ancestry, national origin, religion, sex, sexual orientation, gender identity, age, disability, veteran status, genetic information, marital status or any other legally protected status.

We will ensure that individuals with disabilities are provided reasonable accommodation to participate in the job application or interview process, to perform essential job functions, and to receive other benefits and privileges of employment. Please contact us to request accommodation

The applicable salary range for each U.S.-based role is based on where the employee works and is aligned to one of 4 tiers according to a cost of labor index in that geographic area. Starting pay for the successful applicant will depend on a variety of job-related factors, which may include education, training, experience, location, business needs, or market demands. The expected salary ranges for this role are set forth below. These ranges may be modified in the future.

  • Tier A: USD $212,000 - USD $297,000
  • Tier B: USD $202,000 - USD $282,000
  • Tier C: USD $191,000 - USD $267,000
  • Tier D: USD $179,000 - USD $253,000

You can view which tier applies to where you plan to work here. If your location is not listed, please speak with your recruiter for additional information.

This job is also eligible for participation in Twitter’s Performance Bonus Plan and Equity Incentive Plan subject to the terms of the applicable plans and policies.

Twitter offers a wide range of benefits to U.S.-based employees, including medical, dental, and vision insurance, 401(k) program with employer match, generous time off for vacation, sick time, and parental leave. Twitter’s benefits prioritize employee wellness and progressive support to our diverse workforce.

Location

Seattle, Seattle, San Francisco

 

Application

U.S. Equal Employment Opportunity information (Completion is voluntary)
Non U.S. Equal Employment Opportunity information (Completion is voluntary)
Privacy and data