Earthmover is building a product that solves the biggest storage and compute challenges of our time for users in the deep sciences - think climate, earth observations, biotech, machine learning, and AI.

The Earthmover platform today has two components. Our first product, Arraylake, is a data lake platform for multi-dimensional scientific array data. It enables users to manage and process array data in open standard cloud-native formats, with features like versioning, cataloging, collaboration and access control on top. Our second product, Flux, sits on top of Arraylake and provides an API gateway to multi-dimensional data through standard-compliant APIs. We are also heavy open source contributors in the scientific data space, leading the development of projects such as Zarr, Xarray and our own open source multidimensional array store, Icechunk. We’re actively building powerful features and experiences on top of Icechunk, Arraylake and Flux while also adding more products to our platform to advance the field of scientific computing and accelerate progress in many domains. Some specific problems we’re thinking about:

• Collaboration between scientists, teams and organizations. Getting discoverability, sharing, and access controls right.
• Transparent, high performance array access, along any dimension.
• Groundbreaking open access public datasets to the entire scientific community.
• Generating insights for organizations about how they can optimize datasets on our platform.
• Navigation, exploration, and visualization of nuanced hierarchical scientific data structures.
• Providing highly scalable, highly available, multi-region APIs for data discovery and delivery

Job Description

As a scientific data platform company, our product has two key objectives: provide a best in class array storage and processing system, and deliver an exceptional collaborative user experience on top of this. We’re at a stage in our product development lifecycle where we’ve built a killer foundation, and our focus is now on shipping new features and products that help our users solve challenges unique to scientific data.

Responsibilities

• Design and implement features that allow our customers to quickly extract insights from their data and processes.
• Own our web authentication and authorization layers, and continue to expand access control features for organizations, users, and API keys.
• Own the frontend unit and integration testing process, improving our capability to rapidly ship frontend code with confidence.
• Work on core parts of our web application, integrating with our API’s to continuously improve the user experience and capabilities of our product.

You’re a great fit if you

• Are product minded, and excited to work directly with customers & scientists to shape our product to solve their problems.
• Have experience designing and building high quality user interfaces using React, especially for data visualization and analysis applications.
• Have experience integrating with control layers, user permission systems, or access control tools.
• Would enjoy partnering with other parts of the stack to rapidly iterate on new product features, in particular our service API and client libraries. You should be able to understand the whole stack, from low level cloud performance to usability and user experience concerns.
• Have at least 6 years experience as a software engineer working on frontend or full-stack web applications. Our stack is written in Python, TypeScript, and Rust, experience with any typed backend language is valuable.
• Have a genuine enthusiasm for this job description, even if you don’t yet have experience with all of the listed responsibilities. We have a top class team and are open to helping the right candidate grow into these responsibilities over time.

Our stack

• We’re building Earthmover to be a cloud native data platform for scientific data. We are deployed on AWS and have active Google Cloud and on-prem environments.
• Our front-end application is built on NextJS and is deployed on Vercel.
• Our service and client are both written in Python, leveraging asynchronous interfaces as much as possible (e.g. FastAPI, HTTPX, Motor, Aiobotocore).
• Subsets of our stack are written in Rust, and we’re interested in increasing this footprint over time.
• Our infrastructure as code stack is based on Pulumi and is written in Typescript.