Applied Research Scientist, Machine Learning
San Francisco, CA

Job Description

While more and more people share their little stories that make your day better or declare manifestos that change the world, our recommendation algorithms and our understanding of natural language has to handle new topics, new interests, new styles and new audiences every day.

Research at Medium is about helping our team discover connections between readers and writers, from professional journalists to amateur cooks, shared on a beautiful, collaborative tool. Methods we use span from collaborative filtering approaches, to supervised and unsupervised learning models, from deep learning to natural language processing techniques.

How do we introduce a new article just published by the Economist about Elizabeth Taylor? Are the movie enthusiasts or the jewelry fans the right audience? Is 'origami' the art of paper folding or a new software? Answering these questions will require you to apply well-studied machine learning approaches, and invent your own algorithmic solutions.

We are looking for an Applied Research Scientist in Machine Learning that will work closely with other engineers to build and train models to recommend the articles that our readers will love, and understand natural language in Medium posts better. You will build machine learning model prototypes, use statistical learning tools to discover structure in our data coming from millions of interactions daily.

At Medium, we are proud of our product, our team, and our culture. From frontend to API, from data collection to product management, Medium engineers work multi-functionally with open communication and feedback.

What will you do

* You will select and tune machine learning models and measure their performance for recommendation tasks and many other natural language processing tasks.
* You will work closely with system engineers to get your research ideas into production and run experiments.
* You will use tools to extract and transform data, engineer features for your model, build offline evaluation tools.
* You will apply state-of-the-art machine learning technologies to find the most interesting and relevant for readers on Medium and help them get value from Medium by measuring model performance in offline setting.
* You will collaborate with data scientists, product managers, and other engineers across the company.

Who are you

* You have worked for 2+ years on machine learning models for item recommendation and NLP applications.
* You have experience working on production machine learning systems at scale, analytics, search, ranking, recommendations, and/or natural language processing.
* You have deep understanding of machine learning models, feature engineering, and how to evaluate model performance.

Nice to have

* Python fluency, and familiarity with Python ML libraries, and Py-Spark.
* Experience using information retrieval methods including vector space model, latent semantic indexing and latent Dirichlet allocation.

At Medium, we foster an inclusive, supportive, fun yet challenging team environment. We value having a team that is made up of a diverse set of backgrounds and respect the healthy expression of diverse opinions. We embrace experimentation and the examination of all kinds of ideas through reasoning and testing. Come join us as we continue to change the world of digital media. Medium is an equal opportunity employer.

Interested? We'd love to hear from you.