With Ray, we're making it easy to program at any scale (from your laptop to the datacenter) by providing easy-to-use, general-purpose, and high-performance tools. In addition, we are building a rich ecosystem of libraries (for reinforcement learning, hyperparameter search, experiment management, machine learning training, prediction serving, and more) on top of the core distributed system so that users can rapidly build sophisticated applications. Help us build the future of software development.
We're funded by well-known investors.
Our team is a mix of machine learning and distributed systems PhDs and PhD dropouts from UC Berkeley.
Our goal is to make it easy to program at any scale, and we're not afraid of hard problems.
Right now we are a fairly small engineering team.
Here are some processes we employ:
Code review: every pull request is reviewed by another engineer prior to merge, and must pass unit and integration tests prior to merge.
Design reviews: for larger projects, architecture and high level designs are written up in a design documented and reviewed in a group meeting.
Standups: we have twice weekly standups to discuss progress and blockers.
Product roadmap: we have weekly meetings for the product roadmap, and quarterly retreats for planning at the company level.
Distributed systems: With the Ray distributed system at the core of our product, we face daily challenges with (1) debugging, (2) testing, and (3) designing reliable systems.
Performance: As a lower-level framework, Ray must provide close to bare metal performance for both low latency and high throughput ML workloads.
API design: With Ray and higher-level ML libraries (e.g., Tune, RLlib), we are pushing the state of the art in APIs for distributed machine learning. Designing simple and general APIs for users is a key challenge for the success of Ray and its ML libraries.
Many languages: As a multi-language system, Ray has to work with not only Python but also interoperate with languages such as Java, C, Rust, etc.
We are an early-stage startup that believes in hard work
Most of our engineers are in the office from 10 am - 6 pm. We don't have any hard rules around working hours. Attend your meetings, get your work done, your time is your own!
We provide the best snacks and drinks that you choose! As well as lunch every day.
Anyscale has a flexible PTO policy. Take the time that you need to be recharged and relaxed so you can do your best work.
Your choice of the top medical, dental and vision plans.
From volunteering together to board game nights, we love doing team activities! We have quarterly offsite's, an annual holiday party, happy hours and hangouts sprinkled in-between.
Interested in this company?
Skip straight to final-round interviews by applying through Triplebyte.