The Search and Machine Learning team at Vimeo is focused on developing, deploying, and maintaining machine learning driven solutions at Vimeo. Join us and help bring our feature ideas (and your own!) off the whiteboard and into reality. As a Machine Learning Engineer, you will research improvements in data collection, feature engineering, and algorithmic optimization. You will also work on implementing your models in production systems and data pipelines.
What You'll Do:
Work with high-level applications of Machine Learning:
Video Understanding: Use Deep Learning and Recurrent Neural Networks to identify people, things, and styles within our 300M+ Videos
Text Understanding: Use NLP and Entity Extraction to extract key features from Video descriptions, tags, profiles and album descriptions for use in search
Visual Search: Create feature embeddings that allow users to search for similar stock videos with images and videos (in addition to text search)
Recommendations: Make it easy for users to connect to the best content, creators, and audience
Personalization:Optimize a user's experience based on what we know about their preferences and history
Ranking: Prioritize most important search results, notifications, and news feed items that we should show a given user
Design and build machine learning systems that solve difficult problems involving big data, information retrieval, video understanding, ranking, and recommendation systems
Write, test and maintain production-quality ML models and services at Vimeo scale (200TB of data per month)
Study and analyze problems, propose solutions and design experiments
Skills and knowledge you should possess:
Masters in Computer Science or related technical field or equivalent practical experience
3+ years experience with Python
Strong knowledge of ML techniques including both supervised and unsupervised learning, classification, regression, and optimization
Experience with TensorFlow, PyTorch, SKLearn, XGBoost or similar libraries
Experience delivering Computer Vision, Video Understanding, Personalization and/or Recommendation solutions to users at scale
Experience shipping models into a high availability production environment
Experience with data processing / ETL tools at scale like Spark, HDFS, Apache Beam, etc
Experience optimizing search via Solr / Elasticsearch and/or nearest neighbors via Annoy / FAISS
Go and C / C++
A strong mathematical background (Probability and Statistics)
PhD with publications in top-tier conferences or journals in ML (such as NeurIPS, ICML, CVPR, SIGGRAPH)
At Vimeo, our mission is to empower video creators to tell exceptional stories and connect with their audiences and communities. Home to more than 80 million members in over 150 countries, Vimeo is the world’s largest ad-free open video platform, providing powerful tools to host, share and sell videos in the highest quality possible.
We work hard to enable creators of all kinds to succeed, and to that end, we prioritize attracting diverse talent and cultivating an inclusive environment that encourages collaboration and creativity. We’re committed to building a company and a community where people thrive by being themselves and are inspired to do their best work every day.
Vimeo is based in New York City, with additional offices in Europe and India. Vimeo is an operating business of IAC (NASDAQ: IAC). Learn more at www.vimeo.com.