Principal Software Engineer, Cube

• Expert in database algorithms and data structures: columnar vs row storage, SQL parsing, query planning, indexes, partitioning, join algorithms, filter pushdown optimizations, HLL
• Expert in Rust, or any other low-level language
• Good communication skills
• Fluent English
• Previous startup experience or interest in working in a small fast-moving company

Desirable

• As an engineer of the Cube Core team, your main goal will be to build and maintain various parts of open-source Cube Core (https://github.com/cube-js/cube.js), community support in our Slack, as well as development of Cube Cloud product
• Some of the problems you’ll be working on:
• SQL generation and rewriting. Cube is a simple yet powerful way to organize SQL queries. The Cube way is to generate the best possible SQL query for a given database while abstracting all complex stuff from users and giving them full power of SQL at the same time. To solve this problem, Cube parses SQL queries written by users to understand the structure and rewrites them to apply performance optimizations
• Cube Store. Cube Store is used to serve analytics for trillions of data points with sub-second response times. To keep up-to-date with growing big data demands and serving speeds we’re developing our own database optimized for serving huge aggregated tables with latencies of several milliseconds
• Cube APM. One of the main tech problems for Cube Cloud is to provide APM analytics of Cube instances using Cube itself. Being purely a dog food problem, large-scale real-time analytics has a lot of challenges, most of which will be solved for the first time ever using the SQL approach