Voltron Data

We develop composable data systems that provide flexible, productive, and cost-optimized modular components built on open-source standards. We were founded in 2021 and operate in the Software Development industry. We started Voltron Data to bridge the gap between state-of-the-art SaaS and open source.

We actively maintain and contribute to open-source projects, including Apache Arrow, Ibis, Substrait, RAPIDS, Velox, and Parquet. Our primary product is Theseus, a GPU-accelerated SQL engine for AI workloads.

Theseus orchestrates end-to-end RAG AI pipelines. It is GPU-accelerated, distributed, and AI-native. It parallelizes complex queries and optimizes query execution for GPUs, delivering rapid results. Theseus incorporates GPU-powered User-Defined Functions (UDFs) into SQL. This enables full pipelines, from embedding and vector search to RAG in Dataframes and SQL.

Theseus quickly queries data lakes, lakehouses, and warehouses in place. It supports Apache Iceberg and standard file formats. Its Kubernetes-native control plane deploys across public and private clouds. It supports existing authentication and air-gapped environments for security and compliance.

Theseus also includes a Query Profiler, built for heterogeneous, accelerator-native environments. This tool monitors engine performance at runtime and allows real-time analysis of query performance at CPU, GPU, and I/O layers with zero overhead. You can get Theseus via AWS Marketplace or as an Enterprise Edition.

Our team has between 51 and 200 employees. Our team consists of data system engineers, scientists, analysts, and open-source maintainers. They have a history of driving innovation in data analytics.