Feldera
Feldera provides an incremental SQL computation engine based on DBSP theory that updates query results in real time as data changes, serving data engineering, ML infrastructure, security, and observability teams via an open-source edition and a self-hosted enterprise product.
- Company typePrivate
- Founded2023
- HeadquartersPalo Alto, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What Feldera does
Feldera is a Sunnyvale-based private company that has built an incremental SQL computation engine based on DBSP (Database Stream Processor) theory, a research contribution from its founding team that won the Best Paper award at VLDB 2023. The platform maintains arbitrary SQL pipelines incrementally — updating only the portions of query results affected by input changes rather than recomputing full queries — and is built on Apache DataFusion with a SQL-to-DBSP compiler that generates Rust code. Feldera positions the engine as solving a fundamental problem in modern data architectures: as data volumes and agent-driven workloads scale, traditional batch recomputation becomes prohibitively expensive and slow, while Feldera claims to keep compute costs and latencies flat even at 1000x scale.
The product surface includes an MIT-licensed Open Source Edition for single-node deployments and an Enterprise Edition for self-hosted Kubernetes with multi-tenant isolation, fault tolerance, S3-backed persistence, enterprise SSO, and dedicated support. Supporting interfaces include a Web Console, an fda CLI, a Python SDK, and a REST API, along with connectors to Kafka, Delta Lake, Apache Iceberg, S3, PostgreSQL, NATS, and DynamoDB, plus a dbt-feldera adapter. Feldera targets data engineering teams, ML/AI infrastructure teams (real-time feature engineering and fraud detection), security/authorization teams (fine-grained access policy evaluation), and observability/SRE teams (OpenTelemetry analytics). Named customers and design partners span Microsoft (SQL Server team), Hopsworks, Eviny Fornybar, OSO, and Solana Vibe Station, with the latter processing 100GB+ of daily blockchain data at millisecond latency.
Feldera's business model is a hybrid open-core: a free open-source tier drives developer adoption and community-building, while enterprise deployments are sold via direct field sales with custom pricing gated behind a 'Book a demo' CTA. The company completed a $6 million seed round led by Costanoa Ventures in August 2023, and Ion Stoica (co-founder of Databricks and Anyscale) is listed as an investor/advisor. SOC 2 Type 2 compliance has been achieved, supporting enterprise procurement. Headcount is reported at 1-10, consistent with an early-stage company still building out commercial and engineering capacity.
Feldera firmographics
Firmographics- Name
- Feldera
- Website
- https://feldera.com
- Company type
- Private
- Founded year
- 2023
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- Feldera provides an incremental SQL computation engine based on DBSP theory that updates query results in real time as data changes, serving data engineering, ML infrastructure, security, and observability teams via an open-source edition and a self-hosted enterprise product.
- Ownership category
- akta.pro rank
Feldera industry classification
Industry- Product category
- Real-time Data Infrastructure
- NAICS
- Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Custom Computer Programming Services (541511)
- SIC
- Services-Computer Programming, Data Processing, Etc. (7370), Services-Computer Integrated Systems Design (7373)
- akta.pro industry
- Real-Time / Streaming Data Integration (CDC & Event Streaming)Hardware, Technology & Telecom/Enterprise Platforms & Databases/Data Integration, ETL/ELT & Middleware
Keywords
Where Feldera is headquartered
LocationHeadquarters
- HQ city
- Palo Alto
- HQ country
- United States
- HQ region
- North America
Markets served
Feldera business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Marketing or Sales, Operations
Revenue model
- Open Source Edition: Free incremental computation engine with MIT license targeting developers and individual users for evaluation and non-production use. Single node deployment with full SQL support and community support channels.
- Enterprise Edition: Self-hosted Kubernetes deployment with multi-tenant isolation, enterprise SSO, secrets management, per-pipeline compute isolation, fault tolerance, S3-backed persistence, and dedicated support. Custom pricing based on use case.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Pay-as-you-go | Open source edition - free tier for developers |
| Subscription | Annual | Enterprise edition - production self-hosted deployment |
Go-to-market motion2 records
Distribution channels3 records
Marketing channels6 records
Feldera product offering
Product offeringCore offering
Feldera provides an incremental computation engine that maintains SQL pipelines of any complexity by updating query results in time and compute proportional to the size of the data change rather than the size of all data. Built on DBSP (Database Stream Processor) theory that won the VLDB 2023 Best Paper award, the platform supports arbitrary SQL including joins, GROUP BYs, and recursive queries, and ships as a free MIT-licensed open source edition or a self-hosted enterprise edition for Kubernetes deployments with multi-tenant isolation, SSO, and fault tolerance.
Product overview
Feldera is an incremental SQL computation platform consisting of a core engine and supporting interfaces. The core is the award-winning incremental compute engine (based on DBSP theory) that maintains SQL pipeline results incrementally as input data changes. Distribution options include the MIT-licensed Open Source Edition for single-node use and the Enterprise Edition for self-hosted Kubernetes deployments with multi-tenant isolation. User interfaces comprise the Web Console for interactive development, the fda CLI tool, Python SDK for programmatic access, and a REST API for integration. The product supports connectors for major data systems (Kafka, Delta Lake, Iceberg, S3, NATS, PostgreSQL, DynamoDB) and integrates with Grafana for visualization and dbt via the dbt-feldera adapter. Feldera Skills extends agentic use cases by enabling SQL pipeline management through AI agents.
Differentiator
Problem solved
Functional benefit
Products and services
- Feldera Continuous Analytics Platform An incremental SQL computation platform consisting of an award-winning incremental compute engine that maintains SQL pipeline results incrementally as input data changes, with support for connectors to major data systems including Kafka, Delta Lake, Iceberg, S3, NATS, PostgreSQL, and DynamoDB. Targets enterprise engineering teams that need real-time data freshness on arbitrarily complex SQL workloads.
Quantifiable outcome
- 10x+ reduction in warehouse compute costs
- +4 more outcomes
Companies that use Feldera
Customer profileNamed customers5 records
Segments4 records
Ideal customer profiles4 records
Feldera technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration10 records
Feature5 records
Feldera partnerships and signals
Strategic signalScale indicators5 records
Recent moves7 records
Expansion highlights6 records
Feldera competitors and assessment
Company assessmentDirect peers
- Materialize: Materialize is the closest direct competitor: a SQL stream processing engine built around incremental view maintenance. It targets the same 'continuously updated SQL views' use case Feldera addresses, including real-time analytics, feature engineering, and event-driven applications.
- Rockset: Rockset (acquired by OpenAI) was a real-time analytics SQL database that indexed data for low-latency queries, similar to Feldera's incremental SQL positioning. It competed for the same developer-driven, real-time analytics workloads, particularly for AI/agent infrastructure use cases.
- ClickHouse: ClickHouse is a high-performance columnar SQL database used for real-time analytics, and through its acquisition of Materialize, now directly bundles an incremental SQL engine. It competes head-to-head with Feldera for real-time SQL analytics at scale, including OLAP and observability workloads.
- StarRocks: StarRocks is a real-time analytical SQL engine optimized for sub-second queries on fresh data, including streaming ingestion and lakehouse architectures. It targets the same horizontal 'real-time SQL over changing data' use case Feldera serves.
- Confluent (ksqlDB): Confluent's ksqlDB provides SQL stream processing on top of Apache Kafka, allowing users to define continuous queries over event streams. It directly overlaps Feldera's streaming engine capability and serves the same Kafka-centric data engineering teams.
- Apache Flink: Apache Flink is the leading open-source stream processing framework, supporting stateful computations over data streams. It overlaps with Feldera's streaming and recursive-query capabilities and is a common alternative for real-time data pipeline architectures.
- Decodable: Decodable is a managed stream processing platform built by the original creators of Apache Flink, offering SQL and pipeline tooling over streaming data. It competes with Feldera for enterprises that want a turnkey real-time data platform rather than building on open source directly.
- Apache Pinot: Apache Pinot is an open-source real-time distributed OLAP datastore used for user-facing analytics at scale. It targets the same low-latency, high-ingest, SQL-driven analytics workloads that Feldera's incremental engine is designed to serve.
Broad incumbents
- Databricks: Databricks is a broad lakehouse and data engineering platform with Delta Live Tables, Photon engine, and SQL Warehouse capabilities that increasingly include streaming and incremental processing. It is a 'platform-of-record' competitor Feldera must displace or integrate alongside.
- Snowflake: Snowflake is the dominant cloud data warehouse and has been steadily adding real-time and streaming features (Snowpipe, Dynamic Tables, Snowpark). It is the incumbent Feldera's 10x cost-reduction pitch is largely aimed at displacing or augmenting.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat5 records
Key risks6 records
Key highlights7 records
Customer concentration
Feldera social profiles
Digital presenceFeldera compliance and trust
Trust signalCompliance2 records
Feldera financial estimates
Financial estimateRevenue estimate
Valuation estimate
Feldera leadership team
Management profileNumber of profiles
Profiles5 records
Feldera funding detail
Funding detailFunding overview
Funding rounds2 records
Investors3 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Feldera M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Feldera
What does Feldera do?
Feldera provides an incremental computation engine that maintains SQL pipelines of any complexity by updating query results in time and compute proportional to the size of the data change rather than the size of all data. Built on DBSP (Database Stream Processor) theory that won the VLDB 2023 Best Paper award, the platform supports arbitrary SQL including joins, GROUP BYs, and recursive queries, and ships as a free MIT-licensed open source edition or a self-hosted enterprise edition for Kubernetes deployments with multi-tenant isolation, SSO, and fault tolerance.
Is Feldera a public or private company?
Feldera is a private company. It is classified as venture growth investor backed and is currently operating.
When was Feldera founded?
Feldera was founded in 2023. It employs 11 to 50 people.
Where is Feldera based?
Feldera is headquartered in Palo Alto, United States, in the North America region.
How does Feldera make money?
Two revenue lines are on record. Open Source Edition is the primary driver. The others are enterprise Edition.
Who are Feldera's main competitors?
Direct peers on record are Materialize, Rockset, ClickHouse, StarRocks, Confluent (ksqlDB), Apache Flink, Decodable and Apache Pinot. Broad incumbents are Databricks and Snowflake.
Does Feldera have an API?
Yes. Feldera provides a REST API for pipeline management. Based on changelog documentation, the API exposes endpoints for pipeline lifecycle operations (create, start, pause, resume, stop, delete), monitoring, checkpoint management, and ad-hoc SQL queries. The Python SDK (feldera package) wraps the REST API for programmatic access. Version v0.292.0 introduced pipeline event monitoring via API endpoints. Developer documentation is at docs.feldera.com/api.
What industry is Feldera in?
Feldera's product category is Real-time Data Infrastructure. Its primary akta.pro industry code is HDAEACAB, Real-Time / Streaming Data Integration (CDC & Event Streaming). Its NAICS code is 5182 and its SIC code is 7370.