In the last few years, we’ve seen a push in the data community for lighter-weight tooling. The last generation of data infrastructure (Kafka, Flink, Spark) was designed to handle massive volumes of data. In contrast, a new crop of infrastructure has emerged for companies that have less extreme needs, i.e they don’t have big data. In batch processing, we’ve seen this play out with the rise of single-node query engines like DuckDB.

Meanwhile in stream processing, we have been looking for a team to radically simplify how companies handle real-time data. Stream processing has become essential infrastructure for the biggest companies in the world, powering critical use cases like IoT data processing, real-time ML, live metrics etc. But despite relative ubiquity–Kafka is in use by 80% of the Fortune 100–the standard toolkit for streaming has historically carried a substantial devops burden and steep learning curve.

Enter GlassFlow. Founded by Armend Avdijaj and Ashish Bagri in Berlin, GlassFlow is a new streaming data platform that prioritizes ease of use. GlassFlow is serverless. There are no cluster deployments for customers to manage. GlassFlow is also python-first, which makes it particularly attractive to data and AI engineers. Their API for publishing, consuming, and responding to events can be implemented in minutes. It’s quite likely the simplest way to get started with streaming data pipelines.

Ashish Bagri, Co-Founder & CTO (left), Armend Avdijaj, Co-Founder & CEO (right)

Armend and Ashish are repeat founders who have spent much of their career building data products. They know this problem first-hand and have a clear vision they’re marching towards. We are thrilled to lead a $4.8M seed round and join them on their mission to bring simplicity to real time data processing. Check out their website to learn more, and give them a follow on LinkedIn to stay up to date on future updates!