Feldera Raises $21.5M to Reduce Database Compute Costs by 95% with a Mathematical Breakthrough

Feldera, the incremental view maintenance engine for enterprises, today announced $21.5 million across a Series A round led by Inovia Capital, with participation from Costanoa Ventures and Battery Ventures, and a Seed round led by Costanoa Ventures, with participation from Ion Stoica (co-founder of Databricks and Anyscale). Founded by award-winning computer scientists, Feldera’s engine makes it dramatically cheaper and faster to keep answers fresh over massive datasets, for both AI agents and the people who rely on them.

Enterprises’ Problem: AI Agents Need Fresh Data to Deliver on AI-Ready Products.

Companies today hold more data than at any point in history, yet leveraging that data for AI agents remains painfully expensive and slow. Legacy analytics platforms rely on brute-force execution patterns, re-scanning entire databases every time a query is run, even if only a tiny fraction of the data has updated. This computational redundancy forces an impossible trade-off for AI-ready products: eat skyrocketing compute bills, or accept stale, shallow data that cripples AI and slows operational decision-making.

Feldera’s Solution: Incremental View Maintenance to Reduce Compute Time & Costs

Feldera solves this bottleneck with a breakthrough known as incremental view maintenance that keeps query results continuously fresh across even the most massive enterprise databases. Built on the award-winning theory called DBSP (Database Stream Processing), Feldera automatically converts complex SQL queries into incremental programs that evaluate only new data, eliminating redundant, full re-scans. By plugging directly into existing datalakes without requiring structural re-architecture, the platform slashes compute costs by 95% and more, while transforming hours-long analytical pipelines into sub-second updates. This provides AI agents with continuous, low-latency access to accurate enterprise context without burning valuable tokens or cloud compute on stale results. Customers already run Feldera in production to build live views over massive amounts of data, that in turn help catch fraud in real-time, run billion-dollar logistics operations and power fine-grained authorization for agents.

Key features include:

• Universal SQL Incrementalization: Converts complex SQL programs into incremental programs without falling back to full recomputation.
  
  
• Seamless Integration: Plugs directly into existing enterprise data warehouses, lakes, and pipelines without requiring custom engineering or brittle streaming code.
  
  
• No Rewrites Required: Enterprises can migrate their existing SQL to Feldera with zero code changes, replacing months of migration work.
  
  
• A Fraction of the Hardware: Feldera customers have cut infrastructure costs an average of 10x, and in some workloads, by over 100x.
  
  
• Enterprise Scale, Laptop Simplicity: Feldera processes millions of records per second, fast enough that workloads once requiring a cluster can run on a single laptop
• Mathematically Guaranteed Correctness: backed by DBSP, the theory that proves Feldera’s results always match your warehouse’s, no matter the SQL

“Making timely decisions against massive amounts of fast changing data using traditional methods requires an immense amount of compute,” said Lalith Suresh, CEO and co-founder of Feldera. “AI agents are only as good as the data they can access. So if that data is too stale because it is too expensive to compute, companies will burn tokens on agents that are just going to be wrong all the time.”

Worldclass Research Team Behind 200 Papers

Feldera was founded by five former VMware researchers: Lalith Suresh, Leonid Ryzhyk, Mihai Budiu, Ben Pfaff, and Gerd Zellweger. The team collectively has published over 200 research papers across database internals, distributed systems, and operating systems.

Having repeatedly watched engineers build fragile, ad-hoc workarounds for incremental processing problems, the group united inside VMware Research to solve the issue at its mathematical foundation. That work led to the creation of DBSP, their award-winning research framework for incremental SQL computation.

Inspired by these successes, Feldera was founded to democratize this technology and help engineers solve the toughest data problems without being held back by the limitations of legacy compute infrastructure.‌ Bringing together deep technical credentials from Stanford, Carnegie Mellon, ETH Zurich, UNSW and TU Berlin with decades of hands-on systems design, the team founded Feldera in 2023 to commercialize their breakthrough, turning 50 years of unsolved database theory into an enterprise-ready engine.

“The team at Feldera solved a 50-year-old fundamental database problem with rigorous mathematical proof,” said Taha Mubashir, Partner at Inovia Capital. “As enterprises scale their real-time AI initiatives, the cost and latency of traditional data compute become unsustainable. Feldera provides the foundational compute layer that enables companies to run continuous, complex analytics at a fraction of the cost.”

To try Feldera’s platform today, please visit Feldera.com.

About Feldera

Feldera is the world’s first enterprise-ready Incremental View Maintenance engine. Built on award-winning DBSP database research, Feldera evaluates any SQL program incrementally by computing only on changed data, turning hours-long batch jobs into sub-second results while cutting compute costs by 95% and more. Founded in 2023 by leading researchers in database theory and distributed systems, Feldera is headquartered in San Francisco, CA. For more information, visit Feldera.com

Contact Information

Riley Munks

PR Advisor

Activate PR - https://www.activate-pr.com/

riley@activate-pr.com