Databricks asked for $1 billion. It left the table with five times that, and a valuation to match. The company announced Aug. 14 that it closed a $5 billion funding round at a post-money valuation of $190 billion, led by Coatue, its second raise this year after a July round signed at $188 billion.
The oversubscription tells the story of the AI data-platform market in a single data point: investors wanted in faster than the company wanted to sell. Databricks had signaled a modest raise to fund product development, but demand from large growth funds and sovereign investors swelled the round, according to people familiar with the terms. The company’s revenue run rate has climbed above $7 billion, fueled by the same agentic-AI wave that has lifted every company selling infrastructure for AI applications.
Databricks built its business on a simple premise: companies need a single place to store, process and analyze their data before they can build AI on top of it. The lakehouse architecture the company pioneered combined the cheap storage of data lakes with the structure of warehouses, and that hybrid became the default choice for enterprises running machine-learning workloads. Now those customers want to build AI agents that automate work, and Databricks is positioning itself as the platform where those agents get assembled.
The new capital will go into products that help enterprises build and manage AI applications, the company said. That means expanding the tools that let non-specialist teams orchestrate models, evaluate outputs and connect AI systems to corporate data. Executives described the spend as an acceleration of work already underway, with engineering headcount growing and the company pushing its platform deeper into regulated industries like financial services and health care.
The round’s structure is notable for what it did not include. Databricks had long been expected to go public, and each funding round at rising valuations has fed speculation about an IPO. The company has said little about timing, but the decision to raise again at a $190 billion valuation suggests management sees more private growth ahead. Investors in the round said the company’s metrics justify the price: run-rate revenue above $7 billion, growth in the high double digits, and a margin profile that has improved as the business matured.
The competitive backdrop makes the valuation a bet on execution. Snowflake, Databricks’ principal rival, has its own AI ambitions, and the two companies now fight for the same enterprise data budgets. The market has room for both, analysts said, but the agentic-AI wave rewards whichever platform gets adopted first by the largest companies. Databricks’ customer list, which spans nearly every Fortune 500 industry, gives it an edge in that race.
Coatue’s decision to lead the round at a $190 billion valuation also signals how growth investors now price AI infrastructure. A year ago, a $100 billion valuation for a data-platform company would have seemed extraordinary; today the round closed with multiple funds competing for allocation. The urgency is real, according to people familiar with the process: investors concluded that AI platform companies of this scale are rare, and that waiting for a public listing would mean paying more or getting less.
The company’s own history explains part of the enthusiasm. Founded in 2013 by a team of Berkeley researchers including Ali Ghodsi, now chief executive, Databricks grew out of Apache Spark, the open-source engine that became the backbone of big-data processing. The company turned that technical foundation into a commercial platform and then rode the AI wave by making its system the place where models meet enterprise data. Investors who backed that trajectory early have been rewarded with a run from a roughly $28 billion valuation in 2021 to $190 billion now.
The new money arrives with strings attached in the form of expectations. Databricks said it will continue investing heavily, and the company’s burn rate is meaningful even at its scale. The $5 billion raise extends its runway and funds the agentic-AI push, but it also raises the bar for any eventual IPO: the company will need to justify a valuation that already prices in years of growth.
Databricks’ position at the intersection of data and AI has made it a bellwether for enterprise software valuations. The company discloses its finances sparingly, but the run-rate figure of more than $7 billion, disclosed with the round, is among the highest in private software. Its customers span banks, insurers, retailers and government agencies, and executives said the agentic-AI push has deepened rather than replaced the core data business. Each agent a customer deploys needs the same lakehouse underneath, which means the new product line sells more of the old one.
For the market, the round closes a remarkable stretch for Databricks. Two raises in two months, a valuation that jumped $2 billion between them, and demand that overwhelmed the company’s own plans. The phrase used by one investor in the round captures the mood: the company wanted to raise $1 billion and got $5 billion because the people writing checks were more convinced of the opportunity than the company itself was ready to claim.
