Tinder is seeking world-class machine learning experts to join our engineering team. You will be working with a team of talented engineers and researchers to develop core machine learning algorithms for improving the unique user experience available only at Tinder. You will have tons of responsibility, freedom and an opportunity to have a direct and immediate impact on company growth. We would love to talk to you if you are passionate about helping Tinder users achieve more matches, communicate more effectively, and create more love at a truly global scale.
In this Machine Learning role, you will:
Evolve existing machine learning pipelines and systems in production
Design and implement machine learning algorithms for recommendation and personalization
Develop machine learning models to derive insights and improve product iterations
Create new product features using state-of-the-art machine learning techniques
Work with engineering, analytics and product stakeholders to deliver meaningful impact on cross-functional projects
Some of the problems you'll work on:
Personal Recommendation. The user experience at Tinder is unique and highly personalized. You'll apply a range of algorithms to personalize recommendations in feed, from the latest in Neural Nets, to collaborative filtering, to explore/exploit approaches, such as Multi-armed Bandits, to bring the highest quality dating experience to tens of millions of users.
Anti-Spam. Now that Tinder has more than 100 million users in over 190 countries, we have some unique challenges in protecting the Tinder ecosystem from spammers. How do you model spammer or bot behavior with different intents? How do you detect a botnet stealing Tinder users profile information?
Image Understanding. Tinder hosts hundreds of millions of profile pictures. A better understanding of images will not only help us avoid inappropriate content, but also understand user interests and preferences in absence of explicit user input. The state-of-the-art computer vision techniques, such as Convolutional Neural Nets (CNN), will be applicable here.
Dialogue Understanding. People talk after they match on Tinder. How do you suggest conversation topics to break the ice? Can we detect when a dialogue goes awry? There are huge opportunities for natural language processing in Tinder.
We're looking for:
MS (PhD preferred) in Computer Science, Statistics or related discipline
4+ years programming experience in languages such as Python, Scala or Java
4+ years industry experience in a range of classification and optimization problems including but not limited to recommendation systems, search relevance, fraud detection, spam detection, etc.
Strong track record of delivering critical business impact in cross-functional projects
Expert knowledge with statistical analysis and methods
Experience with Spark Streaming, Spark MLlib, Python NumPy, SciPy, Scikit-learn, and R
Experience with large-scale distributed systems
As part of our team, you'll enjoy:
The hustle of a startup with the impact of a global business
Tremendous opportunity to solve some of the industry's most exciting problems
Working with an extraordinary team of smart, creative, fun and highly motivated people
Comprehensive health coverage, competitive salary, 401(k) match and meaningful equity
Unlimited vacation and flexible working hours
Daily catered lunches/dinners, endless supply of snacks/refreshments, fitness classes, and social events
Modern, uplifting work environment in an ideal location