Software Engineer
Spiral
Software Engineer
Spiral
Open-source columnar DB with emergent schema
Spiral builds a fast analytical database for multimodal, multi-rate data streams, on top of the open source Vortex file format. Our users are AI/ML researchers and AI infra engineers developing models in complex domains, such as weather & climate, financial, time-series, genomics, point-clouds, videos, and images. They spend their days waiting on data loaders, writing video or sensor data pipelines, and watching expensive GPUs sit idle on I/O. We make that pain go away.
We’re looking for exceptional engineers with deep experience in low-level, high-performance software and cloud-scale storage systems.
5+ years of experience writing high-quality production code
Strong foundation in operating systems and systems programming
Proven track record in performance engineering
Familiarity with tools like Apache Arrow, Parquet, DataFusion, Clickhouse, or DuckDB is a plus
Understanding of cutting-edge ML infrastructure stack (e.g. PyTorch, CUDA) is also a plus
Experience with Rust is a bonus
Willingness to work in-person at our NYC, SF, or London office
Spiral builds Vortex, an open-source columnar database format designed to store diverse data types, including large values like images and videos, inside tabular structures. Its emergent schema lets users append columns or rows and enrich objects with new properties as data grows, avoiding rewrites and supporting millions of columns. The system uses a forest of LSM trees and object-storage formats, with features like cell push-down to boost performance and plans to integrate with Apache Arrow and Iceberg for interoperability. Its goal is to provide a high-performance, flexible analytics platform with storage–compute cohesion and a bring-your-own-compute model for scalable, evolving data workloads.
Explore more opportunities at Spiral