Firebolt
Unicorn · $1.4BFounded 2019 · 144 employees on LinkedIn · 9 known investors
Firebolt is a Postgres-compliant analytical database designed for engineers and AI agents, running from a single laptop binary to multi-node cloud clusters. The product targets teams building data-intensive applications requiring sub-second query latency and is built for next-generation data product development.
Also known as Firebolt Analytics · Firebolt Core
Founders & leadership
Firebolt was founded in 2019 by Eldad Farkash and Saar Bitner.
Investors · 9
Also in the syndicate · 1
Valuation · disclosed
Disclosed eventsSource: SEC prospectus filings, and round valuations the company or its investors disclosed — follow each entry's link for the claim.
Company profile
researched Aug 2026Firebolt develops an analytical database aimed at engineering teams building data-intensive and real-time analytics products. The engine uses a PostgreSQL SQL dialect and is distributed as a single binary that runs on a laptop with SQLite for metadata, or as a cluster of up to 256 nodes using Postgres for metadata, with object storage as the underlying substrate. It handles data volumes from gigabytes to petabytes, supports both real-time and batch analytics, and offers ACID transactions with snapshot isolation. Storage uses a custom format for real-time workloads and Parquet plus Apache Iceberg for batch; connectivity is available via ADBC, a JSON REST interface and the Postgres wire protocol.
An open-source version of the engine was in preview at the time of the sources, with general availability announced as forthcoming, alongside a previously released self-hosted edition called Firebolt Core. Deployment options include a raw binary, Docker, Helm charts and a Kubernetes operator, on x86 or ARM, on bare metal, VMs, any cloud or on-premises data centers. Cloud support is generally available on AWS and GCP, with Azure in preview. Recent engine releases have added Iceberg catalog attachment, a native JSON type with user-defined schemas and Iceberg schema evolution, HNSW vector search indexes over float arrays, cross-database joins, and Tableau Cloud connectivity.
Earlier company materials described the product as a cloud data warehouse delivering sub-second analytic experiences at scale for engineering teams, allowing analysis of larger data volumes than previously possible.
Founding story
Firebolt was founded by Eldad Farkash and Saar Bitner, who previously worked together at business intelligence company Sisense, of which Farkash was a co-founder. Farkash served as CEO and Bitner as COO at founding. The company emerged from stealth in early 2021.
Business model
Open core: the engine is open source and can be self-hosted without limits on the user's own hardware, while Firebolt monetizes a managed cloud service and a bring-your-own-cloud option with support, audit logs and metrics.
Consumption-based pricing for the managed service: compute is billed per second by engine size (from $1.84/hour for a single-node, 8-vCPU S engine to $58.88/hour for a four-node, 256-vCPU 4XL engine in US East), with object storage passed through at $0.0264 per GB per month. New users receive $200 in free credits.
Traction
By January 2022 the company reported strong adoption among fast-growing technology companies after roughly a year of operation, with the team nearly doubling in six months to 200 people across 25 countries. The managed service holds SOC 2, ISO 27018, ISO 27001 and HIPAA compliance and operates in six cloud regions.
Latest developments
The open-source version of Firebolt is in preview with a general availability countdown running on the company site. Recent releases added Iceberg catalog attachment via CREATE ICEBERG DATABASE (4.32), a native JSON data type, user-defined schemas, Iceberg schema evolution and HNSW vector indexes (4.31), vector search indexes for low-latency similarity search (4.30), cross-database joins and queries (4.29), and Tableau Cloud connectivity plus role-based transfer of object ownership (4.28). Azure support is in preview alongside generally available AWS and GCP.
▸Full profile — market position, technology, go-to-market, geography, history, risks & controversies
Market position
Positioned as a high-performance analytical database competing on price/performance and deployment flexibility against cloud data warehouses and analytical engines; the company reached a $1.4 billion valuation with its 2022 Series C and states $270M raised in total.
Combines an open-source engine and open storage formats (Postgres-compatible SQL, Apache Iceberg, DuckLake) with a single binary that behaves identically on a laptop and in a large cluster, minimal dependencies, sub-second query latency targets, and support for both real-time and batch analytics.
Technology
Columnar analytical query engine with a planner-runtime-storage pipeline built around object storage. Features include PostgreSQL-dialect SQL, ACID multi-statement transactions using MVCC and OCC, snapshot isolation, zero-copy cloning, data pruning and late materialization, automated column statistics, redundant-join elimination, geospatial support via the S2 Geometry Library and a GEOGRAPHY type, HNSW vector search indexes, automatic cache warmup, zero-downtime live engine upgrades using shadow clusters, sub-second reads of Apache Iceberg tables via READ_ICEBERG, and an MCP server for connecting LLM tools such as Claude and Copilot.
Go-to-market
Self-service developer-led adoption via a one-line install script and free open-source download, free managed-service credits, and documentation and 'ask an engineer' support channels, supported by a technical blog and podcast program; a Chief Growth Officer and CMO are part of the leadership team.
Software and data engineers, and teams building data-intensive applications, real-time analytics and ELT pipelines, including AI agent workloads; the company states its customers include top global tech companies.
Geography
Offices in San Francisco, Munich, Tel Aviv and Bangalore; a Seattle office was announced in January 2022. The managed service runs in US East (N. Virginia), US West (Oregon), Europe (Frankfurt), Europe (Ireland), Asia Pacific (Singapore) and Asia Pacific (Mumbai).
History
Company materials give a founding year of 2019, while press coverage of the Series C states the company was founded in 2018. Firebolt emerged from stealth around January 2021, having raised a $37 million Series A in December 2020, followed by a $127 million Series B in June 2021 and a $100 million Series C in January 2022 at a $1.4 billion valuation, which also coincided with the hiring of CTO Mosha Pasumansky, formerly a principal engineer on Google's BigQuery team, and the opening of a Seattle office. The product later expanded to a self-hosted free edition (Firebolt Core) and an open-source engine in preview ahead of general availability. Benjamin Wagner is listed as CEO, with Farkash and Bitner listed as co-founders.
Risks & controversies
Public sources give conflicting founding years (2019 on the company site versus 2018 in press coverage) and differing total funding figures ($270M stated by the company versus $269M reported at the Series C). Leadership has changed since founding, with Benjamin Wagner listed as CEO rather than co-founder Eldad Farkash.
Compiled by commissioned research from 8 cited public sources — announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed — not independently audited.
Competitors · 10
by search overlapCompanies competing with Firebolt for the same Google search keywords, organic and paid, via search-intersection analysis.
Timeline · 9
launches, deals, and filingsFirebolt's open-source analytical database is available in preview, with a general availability launch countdown published on the company site; the engine installs via a single script and runs as a single binary.
Version 4.32 adds support for attaching Iceberg catalogs with CREATE ICEBERG DATABASE; 4.31 added a native JSON data type, user-defined schemas, Iceberg schema evolution and HNSW vector indexes; 4.30 added HNSW vector search indexes; 4.29 added cross-database joins; 4.28 added Tableau Cloud connectivity.
Firebolt released an MCP Server allowing connection of Firebolt to AI tools such as Claude and Copilot for conversational data interaction.
Firebolt launched Firebolt Core, a self-managed version of the product offered free, discussed by the company's CTO and VP of Engineering.
Alongside its Series C, Firebolt announced the opening of a Seattle office; the team had nearly doubled in six months to 200 people across 25 countries.
Firebolt announced a $100 million Series C led by Alkeon Capital, with Sozo Ventures, Glynn Capital and existing investors Zeev Ventures, Angular Ventures, Dawn Capital, Bessemer Venture Partners, K5 Global and TLV Partners, reaching a $1.4 billion valuation and $269 million total raised.
$100M source ↗
Firebolt announced the hiring of Mosha Pasumansky, previously a principal engineer on Google's BigQuery team, as CTO.
$127M source ↗
$37M source ↗
Dated company events from announcements, filings, and press; legal rows summarize public dockets and regulator releases.
In the news
The Weekly Notable Startup Funding Report: 7/6/26 – AlleyWatchalleywatch.com · Jul 2026▸Research sources · 8
primary sources listed
- Fireboltfirebolt.io · web
8 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Firebolt do?
- Firebolt is an open-source, Postgres-compliant analytical database that runs as a single binary or scales to hundreds of cloud nodes.
- Who founded Firebolt?
- Firebolt was founded by Eldad Farkash, Saar Bitner in 2019.
- Who are Firebolt's investors?
- Firebolt's investors include Alkeon Innovation, Angular Ventures, Bessemer Venture Partners, Dawn Capital, Glynn Capital Management, Sozo Ventures, TLV Partners, Zeev Ventures.













