Open-source data movement platform for ELT pipelines, moving data from APIs, databases and files into warehouses, lakes and AI applications.
jitsucom/jitsu
Open-source event data platform and self-hostable Segment alternative that collects events and routes them to warehouses and SaaS tools.
Open-source alternative to:Segment
- Stars
- 5,112
- Forks
- 402
- Open issues
- 43
- Last commit
- 6 hours ago
What is it?
Jitsu collects event data from websites, mobile apps and servers, then delivers it to data warehouses such as ClickHouse, BigQuery, Snowflake, Redshift and Postgres, plus S3, GCS and a range of SaaS destinations. It is positioned as an alternative to Segment and is MIT-licensed, so the full pipeline can run on your own infrastructure or on the hosted Jitsu Cloud service.
Events can be sent through browser and Node.js SDKs, React bindings, an HTTP API, or a Segment-compatible proxy for migrating existing clients. JavaScript Functions run on each event to filter, transform or enrich it before delivery, and they can be developed in a TypeScript repo with the Jitsu CLI. Connector syncs pull data from third-party sources into a warehouse, and a Live Events view shows events, function logs and destination writes as they happen.
The project is built from several services: a Go ingest endpoint, a TypeScript rotor for routing and functions, a Go bulker for warehouse loading, a Go sync controller, and a Next.js console that also hosts an MCP server for AI agents. It runs on Postgres, Kafka or Redpanda, ClickHouse and MongoDB, with a Helm chart for local Kubernetes setups.
Key features
- 01Collects events from browsers, mobile apps and servers via SDKs, HTTP API or a Segment proxy.
- 02Delivers data to warehouses such as ClickHouse, BigQuery, Snowflake and Postgres, plus many SaaS tools.
- 03Runs JavaScript Functions to filter, transform and enrich events before delivery.
- 04Syncs data from third-party sources into a warehouse using Airbyte-compatible connectors.
- 05Provides Live Events for inspecting events, function logs and destination writes in real time.
- 06Includes an MCP server so AI agents can configure and operate pipelines.
Use cases
- Replacing Segment to cut per-user billing while keeping a compatible tracking API.
- Streaming website and app events into ClickHouse or BigQuery for analytics.
- Enriching or filtering event data with custom TypeScript functions before it reaches downstream tools.
- Self-hosting a complete event pipeline to keep customer data inside your own infrastructure.
Quick start
- 1
Clone the repository
git clone -b newjitsu --single-branch https://github.com/jitsucom/jitsu - 2
Enter the Helm directory
cd jitsu/helm - 3
Start a local Minikube cluster (give it at least 8GB RAM)
minikube start - 4
Deploy the development stack; then run the tunnel command in a second terminal and open http://localhost:3000
./dev-deploy.sh deploy
Things to know
- Self-hosting the quick start requires Kubernetes via Minikube and Helm, plus at least 8GB RAM.
- The full stack is heavy, depending on Postgres, Kafka, ClickHouse, MongoDB and several services.
- The tunnel step needs sudo and must keep running in its own terminal.
Topics
FAQ
- What is jitsu?
- Open-source event data platform and self-hostable Segment alternative that collects events and routes them to warehouses and SaaS tools. Jitsu collects event data from websites, mobile apps and servers, then delivers it to data warehouses such as ClickHouse, BigQuery, Snowflake, Redshift and Postgres, plus S3, GCS and a range of SaaS destinations. It is positioned as an alternative to Segment and is MIT-licensed, so the full pipeline can run on your own infrastructure or on the hosted Jitsu Cloud service.
- Is jitsu free to use? What license does it have?
- jitsu is open source under the MIT license. Check the license terms before using it in commercial products.
- Is jitsu still maintained?
- Actively maintained. The last commit to jitsu was on Oct 21, 2026.
- What can I use jitsu for?
- Replacing Segment to cut per-user billing while keeping a compatible tracking API. Streaming website and app events into ClickHouse or BigQuery for analytics. Enriching or filtering event data with custom TypeScript functions before it reaches downstream tools. Self-hosting a complete event pipeline to keep customer data inside your own infrastructure.
- How do I install jitsu?
- Clone the repository: git clone -b newjitsu --single-branch https://github.com/jitsucom/jitsu Enter the Helm directory: cd jitsu/helm Start a local Minikube cluster (give it at least 8GB RAM): minikube start Deploy the development stack; then run the tunnel command in a second terminal and open http://localhost:3000: ./dev-deploy.sh deploy
- Where can I get jitsu?
- The source code and installation instructions are on GitHub: https://github.com/jitsucom/jitsu
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source Customer Data Platform in Go that collects events from apps and websites and routes them to warehouses and business tools.
A command-line tool that copies data between databases, files and SaaS platforms in one command, without writing code.
Open-source link attribution platform for short links, conversion tracking, and affiliate programs, built with Next.js and TypeScript.
Open-source product analytics platform combining event analytics, session replay, feature flags, experiments, error tracking, surveys and a data warehouse.
Open-source browser-based BI tool for querying, visualizing, and sharing data from many SQL and NoSQL data sources.