Blog – Mozaik

Star on GitHub

1/* The Mozaik blog.
2
3Notes from building an open-source runtime for teams of
4AI agents — architecture, releases, and what we learn
5shipping it. */

Newsletter

For developers who want to learn how to build self-organizing agents.

Mozaik is live on Product Hunt

Mozaik is live on Product Hunt — TypeScript runtime for self-organizing AI agents.

Check it out

From Fixed Agent Workflows to Intelligent Communication

The Highest Level of AI Agent Autonomy

How to Give AI Agents Autonomy in Highly Regulated Domains

Move From Autonomous Agents to Autonomous Teams

How Three Contributors Shipped Mozaik v3.11 to v3.13 in Two Weeks

How Domain-Driven Design Changed My Approach to AI Agent Architecture

Streaming Is Not a UI Feature Anymore

Multi-agent AI debugging is broken. We built a replay system to fix it.

Inference Interception

Build Your Own CLI Agent: A Step-by-Step Guide

Reactive Agents, Typed Event Handlers, and Agent Swarms: What's New in Mozaik

Structured Context, Context Memory, Context Item Generators, and the Agentic Environment

Multi-Perspective Agentic Systems

Why Agent Frameworks End Up As SDK Wrappers - And How To Overcome It

Why a Single Interface for Multiple LLM Providers Is a Bad Design Decision

Building an Autonomous Agent That Can Run Terminal Commands