Anyscale was founded by the creators of Ray, an open source project from the UC Berkeley RISELab.
Anyscale enables developers of all skill levels to easily build applications that run at any scale, from a laptop to a data center.
Anyscale is backed by some of the top investors like A16Z.
Ray aims to provide a universal API for building distributed applications. To achieve this goal requires a distributed system with high levels of performance and reliability. We're looking for engineers with systems software experience that are interested in contributing to the Ray backend.
Developing scalable distributed applications is notoriously difficult, and debugging these applications is even worse. By exposing the right metrics and visualizations, providing high levels of observability into the underlying system and application behavior, and allowing users to take appropriate actions through the product frontend, we can make the experience of developing, running, and debugging software vastly more efficient and delightful.
One of the things that makes Python great is the fact that it has great libraries. Developers can import numpy and pandas and start building powerful applications. But in the distributed setting today, we don’t have libraries. Instead, we have distributed systems like Spark, Horovod, and TensorFlow Serving. These systems cannot easily be composed together and used as elements of a larger application. For example, an online learning system that ingests streaming data, incrementally trains new recommendation models, and then serves recommendations to users will have to stitch together three different distributed systems (one for streaming, one for training, and one for serving) just to build a single application. In the future, people will build these kinds of applications by importing powerful distributed libraries from a rich ecosystem and by composing them together to build new applications. Help us build that ecosystem. This will include libraries for reinforcement learning, hyperparameter tuning, experiment management, model serving, distributed training, and more. Part of this work will be open source as part of Ray.
RLlib is an open-source library for reinforcement learning that offers both high scalability and a unified API for a variety of applications. We're looking for software engineers with existing machine learning experience that are interested in continuing to improve RLlib.
Site Reliability Engineers (SREs) are responsible for keeping all user-facing services and other Anyscale production systems running smoothly. SREs are a blend of pragmatic operators and software craftspeople that apply sound engineering principles, operational discipline, and mature automation to our environments and the Anyscale codebase.
Developer Relations will connect with developers and speak externally about technologies on conference panels, at meet ups, and on blogs. Your work fosters a community of developers integrating with Ray and Anyscale technologies. Externally, you will work on our YouTube channels, in podcasts and trainings, or support channels to troubleshoot and debug technical problems encountered by developers. Internally, you will work with product engineering teams to improve products by conveying feedback from developers, reviewing and contributing to designs, writing production ready code, and testing new features.
By shaping our online presence on Twitter, Facebook, blogs, and other venues, you will help support, grow, and define the Ray community. For Anyscale, you will be writing and presenting about the product, both to internal and external audiences, in a way that influences the desired behavior. We are looking for machine learning enthusiasts with a passion for communication and experimentation who are scrappy executors who can find creative ways to get things done.
We are looking for a recruiter to do whatever it takes to grow and scale our team. Responsibilities will include finding creative ways to source the best candidates, delivering a delightful candidate experience, implementing scalable processes, and using data to constantly refine our processes.