Paperspace provides an elegant GPU compute platform designed to eliminate infrastructure bottlenecks for developers.
Our Gradient product is CI/CD for Machine Learning: A popular toolkit for developing and deploying Deep Learning models. Teams of all sizes use Gradient to iterate faster and collaborate on intelligent, realtime prediction engines.
Paperspace is backed by leading investors including Y Combinator, Initialized Capital and Data Collective.
Our mission is to make cloud computing more accessible through radical simplicity, community-driven technical resources, and straightforward pricing.
Grew 475% over the past 12 months
We're a pure technology company based in NYC solving hard and impactful problems.
We're backed by leading investors and have a great network through Y Combinator.
Paperspace uses a sprint cycle to plan and deploy new features every 2 to 3 weeks. We keep all our source in Github and use Jira to plan features and track them. Our processes are light-weight, and everyone is involved in the planning and testing process. Developers typically own the features they are working on, and are responsible for their success. Developers work together to finish a feature when it is larger than a few days effort. We have daily stand-up meetings where we report our progress. We are a semi-distributed team, with most devs working in our Brooklyn office. Since everyone is contributing so much we try to be flexible around individual schedules and people working remotely when needed.
We have so many interesting technologies we work with, there is usually more than one project that you will be interested in working on. We try to spread these experiences around so people can get to contribute in many different areas and learn a ton. Some past projects have been in optimal streaming protocols, device drivers, parallel cloud orchestration, new frameworks for API development, Chromium integration, client and server Node.Js work, React component development, integration with new deep learning hardware architectures, hypervisor internals, Docker extensions, object file system security, machine learning framework extensions, and Kubernetes operator development, to just name a few. We use a lot of Golang for microservices, Node.Js, React, Python, C/C++, and even C#. We are focused on making our products available through multiple cloud providers and deploying them through Kubernetes. We want to grow a rich developer community around our products so we are interested in having easy-to-use tools and interfaces that are reliable and well documented. Developers are our focus, both machine learning engineers/teams and devops professionals that need to deploy training and inference pipelines quickly.
Here are some past projects we've given newly joining engineers: a) develop an open-source public API/SDK in Node.Js; b) optimize a GPU kernel for a new frame inference technique; c) integrate our machine learning CI/CD pipeline with Github; d) integrate our model training framework with a Ceph distributed file store to streamline transfer learning applications.
An example of a project we have on the horizon: develop a framework for declarative hyper-parameter optimization which integrates with our deep learning model repository.
Our culture is open, respectful, and flexible. Everyone is passionate about being on the team and making a huge impact though our efforts. We challenge each other when needed, but we don't have a hierarchy of people who know more, and those that know less. We respect everyone's opinions, especially during the planning process, and we encourage everyone to contribute. We also enjoy socializing after a major release, or sometimes just having a random lunch all together. Our environment is very informal. You will see that we care as much about each other as we do our product.
Interested in this company?
Skip straight to final-round interviews by applying through Triplebyte.