Tower, a Berlin-based data infrastructure startup, has successfully raised €5.5 million across pre-seed and seed funding rounds to tackle one of the most pressing challenges in AI-powered software development: the operational gap between AI-generated code and production-ready systems.

Founded by Serhii Sokolenko and Brad Heller, both veterans of Snowflake's engineering organization, Tower emerged from firsthand experience with the evolving challenges facing data engineers. Sokolenko brings extensive product management experience from Databricks, Snowflake, Google Cloud, AWS, and Microsoft, while Heller contributed to Snowflake's control plane architecture. This marks the third startup venture for both founders, indicating their deep commitment to solving infrastructure challenges.

The funding round was led by DIG Ventures for the pre-seed and Speedinvest for the seed stage, with additional participation from Flyer One Ventures, Roosh Ventures, Celero Ventures, and Angel Invest. The angel investor syndicate reads like a directory of data infrastructure leadership, including Jordan Tigani (MotherDuck CEO and Google BigQuery founding engineer), Olivier Pomel (Datadog CEO and co-founder), Ben Liebald (Harvey VP of Engineering), and Maik Taro Wehmeyer (Taktile co-founder and CEO).

Tower's core thesis addresses what the founders describe as a fundamental shift in software development challenges. While AI coding assistants have dramatically simplified code generation, the complexity of testing, debugging, deploying, and operating that code in production environments has become the primary bottleneck. This 'last mile' problem represents a significant opportunity as AI tools become more sophisticated and widely adopted.

The platform architecture centers on Apache Iceberg, an open table format that has achieved widespread industry adoption as the standard for analytical storage. This strategic choice ensures compatibility with major data platforms including Snowflake, Databricks, and other leading vendors while preventing vendor lock-in - a critical consideration for enterprise customers managing sensitive data assets.

Tower's approach combines storage and compute capabilities on a unified platform, specifically optimized for AI agents operating on current, company-specific datasets rather than the stale public internet archives typically used for foundation model training. This design enables more accurate and relevant AI-powered applications that can leverage proprietary business data effectively.

Early market validation appears promising, with Tower reporting significant usage metrics within months of launch. The platform has processed over 200,000 runs across more than 30,000 unique applications, while the company's Python SDK has achieved 70,000 monthly downloads. These self-reported figures suggest genuine developer adoption and platform utility.

Customer perspectives reinforce the market need Tower addresses. Gaurav Saxena, director of engineering at Ford Motor Company, highlighted the strategic value of Apache Iceberg while acknowledging the operational challenges it presents. Many data teams lack the specialized skills required for effective Iceberg deployment and maintenance, creating demand for platforms that can abstract away this complexity.

The competitive landscape includes established giants like Snowflake and Databricks, along with numerous emerging data infrastructure startups, all investing heavily in AI-era data engineering solutions. Tower differentiates itself through its specific focus on the post-generation phase of AI development and its multi-tenant architecture designed for rapid integration and iteration rather than enterprise monolith retrofitting.

Speedinvest's Florian Obst, who led the firm's investment, emphasized Tower's multi-tenant architecture as a key differentiator, noting its design for fast integration and rapid iteration from inception rather than adaptation from existing enterprise systems.

Tower plans to deploy the new capital toward expanding its go-to-market team and deepening platform capabilities. The company's bet is that as AI coding tools become more powerful and prevalent, the operational gap between generated and production-ready code will widen, creating increased demand for specialized infrastructure to bridge this divide.

The timing appears strategic as the AI development ecosystem continues evolving rapidly. Tower's founders argue that the industry has fundamentally shifted from a code-writing challenge to a code-operation challenge, positioning their platform as essential infrastructure for organizations seeking to operationalize AI-generated applications at scale.