Databricks overview
Databricks makes a cloud platform that organizations use to store, process and analyze data and to build machine learning models, AI agents and data applications on top of it. The product, sold since 2023 as the Databricks Data Intelligence Platform and presented in 2026 as the Databricks Data + AI Platform, runs on Amazon Web Services, Microsoft Azure and Google Cloud. The company’s founders created Apache Spark at UC Berkeley, and Databricks itself started widely used open-source projects including Delta Lake, MLflow and Unity Catalog.
Spark began in 2009 as a research project by Matei Zaharia at Berkeley’s AMPLab, aimed at running large data jobs faster than Hadoop MapReduce. In 2013 Zaharia and six colleagues, Ali Ghodsi, Ion Stoica, Patrick Wendell, Reynold Xin, Andy Konwinski and Arsalan Tavakoli-Shiraji, founded Databricks in Berkeley to sell Spark as a hosted service, raising $13.9 million from Andreessen Horowitz. Stoica, a Berkeley professor, was the first chief executive. The company moved its headquarters to San Francisco in January 2015, and a year later Ghodsi, then head of engineering and product, took over as CEO.
Databricks first made Spark easier to run and then widened its ambitions. Microsoft launched Azure Databricks as a first-party Azure service in 2018, and Databricks began promoting the lakehouse, an architecture that adds data warehouse features such as transactions, governance and fast SQL to inexpensive cloud data lakes. It moved into generative AI with the purchase of MosaicML for about $1.3 billion in 2023, bought Tabular, the company started by the creators of Apache Iceberg, in 2024, and acquired serverless Postgres startup Neon for about $1 billion in 2025 as the basis for its Lakebase database.
Its private valuation has climbed steeply. In August 2026 Databricks closed a $5 billion funding round led by Coatue at a $190 billion valuation and said its annualized revenue run-rate had passed $7 billion, up more than 80% from a year earlier. Its products now include Lakeflow for data pipelines, Lakehouse for SQL analytics, the Genie AI assistant, Agent Bricks for building AI agents, Lakebase and, since 2026, a security product called Lakewatch. According to the company, more than 20,000 organizations use its platform; Databricks employs more than 10,000 people and remained privately held as of September 2026.
Software by Databricks
3 products profiled, from Apache Spark (2010) to MLflow (2018).
- Databricks Data Intelligence PlatformLakehouse platform for data engineering, analytics, AI and agents Data & AI Platforms 2014 Usage-based
- Apache SparkOpen-source engine for large-scale data processing and analytics Data & AI Platforms 2010 Free and open source
- MLflowOpen-source platform for tracking, evaluating and deploying ML and AI Data & AI Platforms 2018 Free and open source
Databricks history and milestones
- 2013
Seven researchers behind Apache Spark at UC Berkeley found Databricks and raise $13.9 million from Andreessen Horowitz.
- 2014
Databricks launches Databricks Cloud, a hosted Spark platform, and closes a $33 million Series B led by NEA.
- 2015
Databricks moves its headquarters from Berkeley to San Francisco.
- 2016
Ali Ghodsi succeeds Ion Stoica as CEO, and Stoica becomes executive chairman.
- 2018
Azure Databricks becomes generally available on Microsoft’s cloud, and Databricks introduces MLflow.
- 2019
Databricks releases Delta Lake, an open-source storage layer that brings transactions to data lakes.
- 2021
A $1.6 billion funding round in August values Databricks at $38 billion.
- 2023
Databricks acquires MosaicML for about $1.3 billion and introduces the Data Intelligence Platform.
- 2024
Databricks buys Tabular, open-sources Unity Catalog and announces a $10 billion round at a $62 billion valuation.
- 2025
Databricks acquires Neon, launches Lakebase and Agent Bricks, and announces a Series L round at a $134 billion valuation.
- 2026
Databricks closes a $5 billion round at a $190 billion valuation and reports a revenue run-rate above $7 billion.
Who founded Databricks?
Ali Ghodsi
Co-founder; CEO since 2016
Ali Ghodsi earned his PhD at KTH Royal Institute of Technology in Stockholm and co-founded the peer-to-peer data transfer company Peerialism before arriving at UC Berkeley as a visiting scholar in 2009. There he worked on distributed systems research that produced Apache Mesos and the scheduling method known as dominant resource fairness, and he contributed to Spark and Spark SQL.
At Databricks he ran engineering and product until January 2016, when he succeeded Ion Stoica as chief executive. He has since led the company’s expansion from hosted Spark into data warehousing, databases and AI, and he is also an adjunct professor at Berkeley.
Ion Stoica
Co-founder; CEO from 2013 to 2016, now executive chair
Ion Stoica is a Romanian-American computer scientist who earned his PhD at Carnegie Mellon University and has been a professor at UC Berkeley since 2000, where he co-directed the AMPLab that produced Spark. His research spans networking, peer-to-peer systems such as Chord, and scheduling for large computing clusters.
Stoica served as Databricks’ first chief executive until January 2016, when he became executive chairman. He also co-founded the streaming analytics company Conviva and Anyscale, the company behind the Ray distributed computing framework.
Matei Zaharia
Co-founder and CTO; creator of Apache Spark
Matei Zaharia started Spark in 2009 as a PhD student at UC Berkeley, building it as a faster alternative to MapReduce, and his dissertation on large-scale computing won the 2014 ACM Doctoral Dissertation Award. A Romanian-Canadian graduate of the University of Waterloo, he has been Databricks’ chief technology officer since the company’s founding.
Zaharia also led the creation of MLflow and has kept an academic career alongside the company, teaching at Stanford before joining the UC Berkeley faculty in 2023. In April 2026 the ACM named him the recipient of its 2025 Prize in Computing for his work on data and machine learning systems.
Patrick Wendell
Co-founder and VP of engineering
Patrick Wendell worked on Spark in UC Berkeley’s AMPLab and is one of the project’s founding committers and a member of its project management committee at the Apache Software Foundation. He was Databricks’ director of engineering until January 2016, when he became vice president of engineering in the leadership change that made Ali Ghodsi CEO. In recent years he has overseen the company’s AI infrastructure and built-in AI assistant work.
Reynold Xin
Co-founder and chief architect
Reynold Xin began working on Spark as a doctoral student in UC Berkeley’s AMPLab, where he built Shark, an early SQL engine on Spark that was later replaced by Spark SQL, and GraphX, which became Spark’s graph processing library.
As a Databricks co-founder and chief architect, he started the DataFrame API, the Project Tungsten execution engine and Structured Streaming, all of which became core parts of Spark, and he was release manager for Spark 2.0. In 2014 he led the Databricks team that used Spark to set a world record in the Daytona GraySort benchmark.
Andy Konwinski
Co-founder; later co-founded Perplexity and Laude
Andy Konwinski studied at the University of Wisconsin–Madison and earned his PhD at UC Berkeley, where he helped develop Apache Mesos and Apache Spark alongside Matei Zaharia. In Databricks’ early years he worked much like a product manager, building the Spark community, organizing user conferences and helping the first customers deploy the software.
Konwinski stepped back from day-to-day work at Databricks in 2019. He co-founded the AI search engine Perplexity in 2022 and the venture firm Laude Ventures in 2024, and in 2025 he pledged $100 million to establish the Laude Institute for computer science researchers.
Arsalan Tavakoli-Shiraji
Co-founder and SVP of field engineering
Arsalan Tavakoli-Shiraji holds a bachelor’s degree in engineering from the University of Virginia and a PhD in computer science from UC Berkeley, where he specialized in networking and distributed systems. Before joining Databricks he was an associate principal at McKinsey & Company, advising companies, technology vendors and public-sector clients on cloud computing and IT strategy. At Databricks he is senior vice president of field engineering, leading the technical teams that work with customers adopting the platform.
How Databricks makes money
Databricks charges for usage rather than per user. Customers consume Databricks Units, a normalized measure of processing, billed at per-second granularity at rates that vary by product and workload, such as data engineering jobs, SQL warehouses or model serving. On classic compute, customers also pay their cloud provider for the underlying virtual machines, while serverless products fold that infrastructure into the Databricks bill. Organizations can pay as they go or sign committed-use contracts that trade a minimum level of spending for discounts.
The company sells directly, through cloud marketplaces and, in the case of Azure Databricks, through Microsoft, which offers it as a first-party Azure service. As a private company, Databricks discloses only selected figures. In August 2026 it reported a revenue run-rate above $7 billion, including more than $1.5 billion from its Lakehouse data warehousing product, positive adjusted free cash flow over the prior 12 months and more than 1,000 customers each spending at an annualized rate above $1 million.
Who owns and runs Databricks?
Co-founder Ali Ghodsi has been Databricks’ chief executive since January 2016, and Ion Stoica, the company’s first CEO, is executive chair. Co-founder Matei Zaharia is chief technology officer and David Conte is chief financial officer. Databricks is privately held by its founders, employees and investors, including Andreessen Horowitz, NEA, Thrive Capital, Insight Partners and Coatue, which led its August 2026 funding round. Its board includes Ben Horowitz and NEA general partner Pete Sonsini.
- Ali GhodsiCo-founder and CEO
- Ion StoicaCo-founder and executive chair
- Matei ZahariaCo-founder and chief technology officer
- Reynold XinCo-founder and chief architect
- David ConteChief financial officer
Databricks: frequently asked questions
Who founded Databricks?
Databricks was founded in 2013 by seven UC Berkeley researchers behind Apache Spark: Ali Ghodsi, Ion Stoica, Matei Zaharia, Patrick Wendell, Reynold Xin, Andy Konwinski and Arsalan Tavakoli-Shiraji. Stoica was the first CEO, Zaharia became chief technology officer, and Ghodsi has led the company as CEO since 2016.
Who owns Databricks?
Databricks is a privately held company owned by its founders, employees and investors. Andreessen Horowitz led its first funding round in 2013, and later backers include NEA, Thrive Capital, Insight Partners, Coatue, T. Rowe Price, GIC and Temasek. The company had not gone public as of September 2026.
Who is the CEO of Databricks?
Ali Ghodsi is the CEO of Databricks. A co-founder and former UC Berkeley researcher, he took over as chief executive in January 2016 from fellow co-founder Ion Stoica, who became executive chairman, after serving as the company’s head of engineering and product.
Where is Databricks headquartered?
Databricks is headquartered in San Francisco, California. The company started in Berkeley, close to the university lab where its founders built Apache Spark, moved its headquarters to San Francisco in January 2015 and now has more than 30 offices around the world.
What software does Databricks make?
Databricks makes the Databricks Data Intelligence Platform, now marketed as the Databricks Data + AI Platform, which includes Lakeflow for pipelines, Lakehouse for SQL analytics, Unity Catalog for governance, the Genie AI assistant, Agent Bricks and the Lakebase Postgres database. Its founders created Apache Spark, and the company started MLflow and Delta Lake.
Is Databricks a public company?
No. Databricks is a private company that has financed its growth through funding rounds rather than a stock market listing. Its most recent round, a $5 billion investment led by Coatue that closed in August 2026, valued the company at $190 billion.
Sources
- Databricks surpasses $7B revenue run-rate and closes $5B round, press release databricks.com
- Databricks founders databricks.com
- Databricks announces changes in leadership team, 2016 databricks.com
- Databricks moves HQ to San Francisco, 2015 databricks.com
- Databricks, Wikipedia en.wikipedia.org
Last reviewed . Spotted an error? Send a correction.