Analysis
Databricks has closed the books on its latest mega-round: $5 billion at a $190 billion valuation, led by Coatue, according to CNBC.
It's the culmination of a process Pulse first flagged when talks around a slightly lower private mark surfaced in July -- what's changed since then is that the round is now closed, the valuation firmed up a bit higher, and the investor list has expanded well beyond the initial lead. The roster is notable for its breadth: Blackstone, MGX, and accounts advised by T. Rowe Price joined alongside new investor Sixth Street Growth, plus BOND, Clearlake Capital, Point72, Premji Invest and TPG, according to TechCrunch. Existing backers Andreessen Horowitz, Dragoneer, Goldman Sachs Alternatives and Thrive Capital also re-upped. That's a mix of growth-equity, sovereign-adjacent (MGX is Abu Dhabi-backed), and crossover public-market investors all competing for the same allocation.
Databricks, founded in 2013 by a team out of UC Berkeley's AMPLab including Ali Ghodsi, Matei Zaharia and Ion Stoica, built its business on the 'lakehouse' architecture that combines data warehousing and data lake functionality -- a direct challenge to Snowflake, its closest public comparable, which trades at roughly a tenth of Databricks' new private valuation despite having gone public years earlier. The company said it crossed a $7 billion annualized revenue run-rate last quarter, growing more than 80% year-over-year, and has kept adjusted cash flow positive for 12 consecutive months, a profitability marker most AI infrastructure companies at this valuation can't yet claim.
“The roster is notable for its breadth: Blackstone, MGX, and accounts advised by T.”
The more revealing detail is TechCrunch's reporting that Databricks originally intended to raise just $1 billion, and investor demand pushed the number as high as $15 billion in early conversations before the company settled on the round size that ultimately closed.
That's the same dynamic playing out across the AI infrastructure stack right now -- more capital chasing a shrinking list of companies with defensible revenue growth than there is company appetite to absorb it, which is also fueling why deals like Anthropic's prospective mega-valuation IPO ask aren't drawing much pushback despite steep multiples.
The risk sitting underneath the headline valuation: a $190 billion mark on $7 billion of run-rate revenue is roughly 27x, well above where Snowflake or Palantir trade publicly, and it assumes Databricks' 80%-plus growth rate holds as the AI infrastructure market matures and hyperscalers build more of this functionality natively. Watch for whether Databricks uses this capital for M&A -- the company has been an active acquirer of smaller AI tooling startups -- or banks it purely as a pre-IPO cushion.
