Paperspace, a promising IaaS (infrastructure-as-a-service) startup and provider of artificial-intelligence tools for developers, is determined to bring easy-to-use AI functionality to software development.
The Brooklyn, N.Y.-based company has launched a new suite of dev tools called Gradient, which enables the use of AI functionality by any developer—no matter what level of skills he or she may have. Using Gradient, developers can tap into a dedicated cloud-based library of AI solutions for building, training and deploying machine-learning applications.
Gradient is designed to make developing cloud AI as simple as building a modern web service, the company said. Featuring one-click Jupyter notebooks, a GPU (graphical processing unit)-powered job runner and a new integration scheme that lets users add a mere one line of code to run any app on a full GPU cloud, Gradient works to turn basic infrastructure into building blocks for intelligent applications.
The Idea: To Accelerate New AI/ML App Development
The mission behind Gradient is to accelerate the development of new machine-learning products and services across a host of industry verticals, Paperspace said.
“At the high level, we are a front end to GPU compute,” CEO and co-founder Dillon Erb told eWEEK. “What we do at Paperspace is make it easy to get up and running with production-level applications that are centered around GPU computing.
“The killer app for this is in training deep-learning models. Even (GPU chipmaker) Nvidia as a company has transitioned substantially to the AI and machine-learning space (from the gaming industry), and that’s where most of our customers are today, and the tools that we’re building out are for that audience.”
AI is the future of every industry, but today’s complex cloud ecosystem coupled with a massive skills gap are threatening innovation, Paperspace co-founder Daniel Kobran said in a media advisory.
“You shouldn’t need decades of experience to build intelligent applications. We designed Gradient to remove infrastructure bottlenecks so that any developer can build a production-ready cloud AI pipeline,” Kobran said.
“Our mission is to make every developer an AI developer.”
Citizen Development? No, but Close
Does Paperspace’s platform amount to a form of citizen development? That would be a reach, although the toolset ostensibly is fairly easy for a junior-level developer to parse and use.
Gradient gives developers the ability to launch a GPU-backed Jupyter notebook for Python from anywhere. The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and narrative text. Uses include data cleaning and transformation, numerical simulation, statistical modeling, data visualization, machine learning and others.
In just a few clicks in Gradient, a developer can begin training and deploying models by using any of the popular frameworks, such as TensorFlow or PyTorch. Jobs can easily be versioned and reproduced and can run in parallel for rapid iteration, the company said.
Gradient includes additional enterprise-ready services, such as data integrations, user management and access controls, and the ability to orchestrate across cloud providers — all designed to give any company access to a comprehensive platform for building AI applications, Erb said.
Thousands of Organizations Using It Now
Erb said Paperspace already powers tens of thousands of organizations with turnkey virtual machines for performance-intensive applications such as design, visualization, machine learning and artificial intelligence.
Paperspace is a Y Combinator Company. As of 2017, Y Combinator—a Mountain View, Calif.-based venture capital investment incubator—had invested in about 1,450 companies, including Dropbox, Airbnb, Coinbase, Stripe, Reddit, Instacart, Twitch, Cruise Automation, Optimizely, Zenefits, Docker, DoorDash, Mixpanel, Heroku, Machine Zone, Weebly and Paribus. The combined valuation of YC companies has surpassed $80 billion.
For more information or to get started using Gradient, go here.