Databricks chief Ali Ghodsi, in a recent interview with Axios, candidly credited media leaks for creating a frenzy among investors. “The press leaks it, and then your phone explodes with investors,” he added, saying “Suddenly, we saw $19 billion of interest, and I almost fell off the chair.”
This development comes in the backdrop of Databricks—the AI and data analytics company—raising $10 billion in its latest funding round far surpassing the $3-4 billion it initially set out to raise.
“We started small, looking to raise around $3-4 billion, but once the press leaked the news, suddenly all the investors came in,” said Ghodsi. He added that his phone exploded with investors reaching out to him, many of whom he hadn’t spoken to in years. “Investors were like, ‘Ali, it was seven years ago I talked to you, but I love you. Let’s talk,’” quipped Ghodsi.
A month ago, AIM reported that Databricks was planning to raise $1 billion from Thrive Capital. Around the same time, The Information published a report stating that Databricks was looking to raise $7 billion to $9 billion, with a valuation expected to exceed the previously reported $55 billion, reaching approximately $61 billion.
A Historic Funding Round Ever
The company said that it plans to use its latest funding for AI development, acquisitions, global expansion, and employee liquidity. In 2024 alone, the company has made four acquisitions, including the $1 billion acquisition of Tabular in June.
In October, Databricks signed a five-year deal with Amazon to use Trainium chips for building AI models with MosaicML.
Databricks acquired MosaicML in July last year and later used its technology to launch the Databricks Data Intelligence Platform.
Databricks’ Data Intelligence Platform includes AI solutions to support the entire ML lifecycle. Its core offering, Mosaic AI, simplifies building, deploying, and managing AI and ML models. The platform supports generative AI with large language models (LLMs) using techniques like prompt engineering, RAG, fine-tuning, and pretraining.
In March 2024, Databricks launched DBRX, a transformer-based model with 132 billion parameters and 36 billion active during inference. Its MoE architecture outperforms open-source models and rivals closed-source models like GPT-3.5 and Gemini 1.5 Pro.
Snowflake, Who?
This funding comes in light of the escaping rivalry between Snowflake and Databricks.
Recently, Snowflake CEO Sridhar Ramaswamy shared his thoughts on the total cost of ownership (TCO) comparison between Snowflake and what he referred to as “Spark-based SaaS” solutions – a jibe at Databricks.
“Snowflake consistently outperforms Spark-based SaaS with a 30% price-performance improvement… helping teams focus on innovation, not complexity,” Ramaswamy remarked, fueling a heated debate, particularly as Databricks supporters argue that the additional management controls in Spark are essential for customisation.
Interestingly, Databricks subtly acknowledged a shift in strategy, hinting at an aggressive corporate approach. “We had a program called Snow Melt to go after Snowflake—but that’s behind us now,”said Ghodsi.
Databricks has recently enhanced its SQL and business intelligence capabilities, venturing into Snowflake’s traditional domain.
On the other hand, Snowflake has launched products to compete with Databricks in areas like data engineering and machine learning. The company also has an initiative called ‘SparkAttack’ to capture machine learning workflows from Databricks.
Interestingly, Snowflake’s annual revenue run rate is $3.77 billion, higher than Databricks’ $3 billion.
However, in the recent Cerebras Valley event, Ghodsi said that there was a time when Snowflakes used to keep him up at night, but not anymore.
Ghodsi also mentioned that it was quite competitive when they were on their way to acquire Tabular. “Multiple vendors were very interested – I mean, you’re good at guessing who was really interested in that,” subtly indicating the involvement of their biggest rival.
IPO Takes a Backseat
Ghodsi said that the company has purposely delayed the plans of going public, partly due to economic uncertainties like inflation and the election year. “Six months ago, we decided it would be ‘dumb’ to IPO this year. We wanted to wait for stability, especially given the pressure from interest rates and inflation,” he said.
Despite the delay in an IPO, the rapid success of this funding round signals Databricks’ continued growth. Ghodsi added that the company’s primary goal was to provide liquidity to employees who might otherwise have to wait years for an IPO. “The earliest theoretical possibility of an IPO would be next year, and there’s a lockup period, so it would have been too long for employees to wait,” he said.
The post When Ali Ghodsi Almost Fell Off the Chair appeared first on Analytics India Magazine.
