Start the platform
Run one command to start Kafka, Schema Registry, Kafka Connect, governance, and the self-service portal locally.
Create topics, build streams, and connect AI workflows. No Java.
Installation, configuration, API reference, KSML guides.
One-command local setup. Full platform.
Open-source repos, KSML, community contributions.
Public projects, deployment examples.
Step-by-step guides for common tasks.
Deep-dives on Strimzi, KSML, Kafka Streams.
Support for existing users.
Build streams and execute Kafka operations with Axual AI assistant.
Run one command to start Kafka, Schema Registry, Kafka Connect, governance, and the self-service portal locally.
git clone https://gitlab.com/axual public/axual-platform-compose.gitcd axual-platform-composedocker compose up -dCreate a schema, then create a topic and attach the schema. Configure topic settings such as partitions and retention period.
Create a Kafka Connect application, such as LogSource, to generate messages and publish them to your topics in under 5 minutes.
Lost? Ask Axual MCP Server it to guide you through setup or perform tasks for you.
Upload and version AVRO, Protobuf, or JSON Schema formats, with automatic compatibility checks and clear schema ownership per team.
streams:
transactions
topic: payments
keytype: string
valueType: avro
pipelines:
fraud-check:
from: transactions
via:
-type: filter
if: value.amount > 10000
to: alerts
Enforce ownership, access control, and full audit trails on every topic and schema change, plus optional field-level data masking for sensitive data.
Authenticate via OAuth2, SAML, LDAP, or mutual TLS, with role and attribute-based access control (RBAC/ABAC) enforced on every topic. OAuth support now extends across all data plane components, including the schema registry.
Connect databases, APIs, and legacy systems through pre-built connectors (JDBC, REST, and more) configured, not custom-coded, and managed the same way as any other Kafka application.
PostNL built their entire parcel notification system with one person using KSML. No Java. Full production.
Norsk Helsenett streams clinical data across Norway's entire healthcare network. Sub-100ms latency. Zero-trust.
TenneT and Alliander run real-time grid management on Axual. Thousands of meter readings per second, governed end to end.
Sovereign infrastructure for national-level data flows. On-premises. No US vendor dependencies.


.avif)





.avif)





.avif)



Axual is a self-service and governance platform for Apache Kafka aimed at developers and platform engineers. It combines Kafka management, governance, and application development in one platform, so teams can create topics, build streaming apps, and connect AI workflows without writing Java, filing tickets, or waiting on a procurement cycle.
The self-service portal is available at http://localhost:8080/login/local after you start the platform with Docker Compose.
Yes. You can try the full Axual platform on your own laptop by running it locally with Docker Compose, no sign-up, sales call, or procurement required. One command starts Kafka, Schema Registry, Kafka Connect, governance, and the self-service portal locally.
Clone the repository and start it with Docker Compose in three steps:
git clone https://gitlab.com/axual/public/axual-platform-compose.git
cd axual-platform-compose
docker compose up -d
This brings up Kafka, Schema Registry, Kafka Connect, governance, and the self-service portal on your machine.
No. Running the full platform locally with Docker Compose requires no permissions, no sales call, no sign-up, and no procurement process. The platform is designed so developers can run, build, and validate immediately without depending on platform teams.
Yes. KSML (Kafka Streams Markup Language) lets you build Kafka Streams applications using YAML and Python instead of Java. This is how teams without Java developers build production stream processors on Axual.
KSML stands for Kafka Streams Markup Language. It lets you define Kafka Streams pipelines in YAML — filtering, transforming, and routing data, and generating realistic test data — and deploy those applications directly through Axual, all without writing Java.: You can build production-ready streaming applications with KSML in under 20 minutes using YAML and Python, without Java.
You create topics yourself through the self-service portal, within platform guardrails — no ticket to a platform team required. You can also configure topic settings such as partitions and retention period, register schemas, and manage access from the same portal.
Yes. Axual provides self-service topic creation, so you create topics yourself within governance guardrails instead of waiting on the platform team. The same applies to schemas and access requests.
Axual enforces governance through ownership groups, environments (dev, staging, production), schema validation, access requests, and audit trails. All actions run through Axual's Self-Service APIs, so governance, ownership, and approval workflows stay enforced automatically rather than depending on manual checks.
Axual adds governance, security, and a developer portal on top of Strimzi. It addresses the point where do-it-yourself open-source Kafka setups become hard to manage at scale by layering management, governance, and self-service onto your existing foundation.
Axual provides pre-built connectors for databases, APIs, and legacy systems, including JDBC, REST, JMS, and more. You configure them through the connector UI and can deploy a Kafka Connect application to start generating and publishing events.
Use a Kafka Connect application such as LogSource to generate messages and publish them to your topics, which can be done in under 10 minutes. You can also produce realistic test data through KSML or by asking the Axual MCP Server in natural language.
The Axual MCP Server connects AI assistants to Axual so you can execute Kafka operations through natural language. It combines large language model reasoning with Axual platform workflows — reading environments, topics, and deployments, then using MCP tools to carry out actions through the Self-Service APIs while respecting ownership, groups, and access policies.
Connect in three steps: get your MCP server URL from Axual Self-Service, add it to your AI client, then start asking questions. The command to add it is:
claude mcp add --transport http axual [your-mcp-server-url]
Yes. It enforces the same permissions and boundaries as the user, never accesses Kafka directly, keeps data inside your platform, and integrates with your existing Identity and Access Management system. Every action is validated and governed through the Self-Service APIs, including ownership groups and approval workflows.