The numbers behind the platform
Databricks said in February that revenue had grown more than 65% year over year and passed a $5.4 billion run rate. The company also announced more than $7 billion in financing: $5 billion in equity at a $134 billion valuation and $2 billion in debt.
The funding supports a broad product surface spanning lakehouse infrastructure, governance, analytics, model training, serving, and enterprise agents. That breadth is deliberate. Databricks wants customers to move from raw data to AI applications without leaving its control plane.
Why it matters
Most enterprise AI projects fail in the seams: permissions, data quality, evaluation, lineage, and deployment. A platform that already governs the underlying data can reduce those handoffs. It can also bundle aggressively and make standalone tools harder to justify.
The counterweight is complexity. Broad platforms can become difficult to operate, and customers may resist placing data, models, and applications with one vendor. Open formats help, but practical portability is the real test.
What to watch
Track AI product revenue separately from core data workloads, the adoption of agents in production, and customer movement between Databricks and competing clouds. Watch whether open ecosystem promises survive commercial pressure.
The maniacal take: the enterprise AI winner may not own the most famous model. It may own the governed path from company data to a useful decision.
Sources & further reading
Reporting is based on company announcements and attributed coverage. Analysis and interpretation are Maniacal’s own.