Berlin-based Tower has raised $6.4 million in pre-seed and seed funding to reshape data engineering in the AI era, with DIG Ventures leading the pre-seed and Speedinvest leading the seed round, as Vestbee was informed.
- Launched in 2025 by Serhii Sokolenko and Brad Heller, Tower develops a platform for managing data pipelines, analytical storage, and AI-powered data agents. The system allows engineers to orchestrate Python-native workflows, run ETL processes, develop dbt Core models, and deploy data agents within a single environment.
- It integrates storage and compute capabilities through a managed lakehouse built on the Apache Iceberg format, which ensures compatibility with Snowflake, Spark, and other major data engines.
- Users can operate pipelines on either self-hosted runners for sensitive data or serverless infrastructure, while unified logs, metrics, alerts, and scheduling provide oversight and reliability.
- Tower also supports integration with open-source tools like dltHub, Polars, DuckDB, and AI frameworks such as OpenAI, LangChain, and HuggingFace, enabling automated data processing and AI-driven workflows.
Details of the deal
- The round was led by Speedinvest and DIG Ventures, with support from Flyer One Ventures, Roosh Ventures, Celero Ventures, Angel Invest, and angels including Jordan Tigani, Olivier Pomel, Ben Liebald, and Maik Taro Wehmeyer.
"AI has made it easier to write data pipelines, but getting them to run properly in production is still hard - and only getting harder. Serhii and Brad have lived this problem first-hand, and we’re excited to be backing this talented team as they build Tower to tackle such a huge problem," explains Melissa Klinger, Partner at DIG Ventures.
- Tower will use the $6.4 million to grow its team and improve its platform so that AI-generated code and data pipelines can be reliably turned into working systems. The funding will also support integrated storage and processing tools, helping engineers manage company-specific data more efficiently and reduce operational complexity.
