Investing in Eventual

The AI boom is colliding with a harsh reality: the infrastructure powering modern applications wasn’t built for the data fueling them. Every team building with AI, whether it’s to analyze video footage, extract meaning from PDFs, or train models on web-scale image datasets, is facing the same bottleneck. Traditional data engines like Spark and Snowflake weren’t designed for unstructured, multimodal data. And retrofitting them isn’t working.

Engineers shouldn’t have to spend months wrangling infrastructure nor should production data workloads break unexpectedly.

That’s why we’re backing Eventual.

Eventual is creating a new foundation for AI infrastructure that’s purpose-built for multimodal data. Eventual offers Daft, an open-source, high-performance, Python-native engine that unifies processing across modalities. Daft allows developers to query petabytes of images, video, audio, and text just like tabular data. It’s already helping users at Amazon, Together AI, and CloudKitchens move from infrastructure headaches to production-grade applications in record time.

What makes Eventual’s approach so compelling is their inversion of execution logic: the platform natively handles calls to AI APIs, embeddings, vector stores, and the unpredictable chaos of real-world data while maintaining the simplicity and predictability of declarative semantics and seamlessly integrating with existing data storage solutions. While legacy engines buckle at scale, Daft turns messy multimodal workloads into a first-class, fault-tolerant experience that handles retries, memory errors, and external dependencies as core features, not edge cases. Since launching in 2022, Eventual has shipped a production-ready engine already processing petabytes of multimodal data daily. 

None of this would be possible without the remarkable team behind it.

Sammy Sidhu and Jay Chia have spent their careers on the frontlines of multimodal AI infrastructure at DeepScale (acquired by Tesla), Lyft, and Freenome. They’ve lived the pain of broken systems and have assembled an incredible team of distributed systems veterans from Databricks, AWS, GitHub Copilot, and more. 

We’re thrilled to lead Eventual’s Series A round and support Sammy, Jay, and the Eventual team as they transform the data layer for the AI era. As the world shifts from tabular to multimodal, Eventual will help define what comes next.

Join the Eventual community! You can follow Eventual on X and join their Slack. They are actively hiring, so check out opportunities here!

‍