13 stories
We finished 11th on Product Hunt with organic votes — and the thoughtful questions about observability, cost, loops, and human control confirmed Mozaik's roadmap toward controlled autonomy.
Fixed agent workflows break when reality diverges from the happy path. Mozaik replaces predefined graphs with an event-driven runtime — and agents that decide when and with whom to communicate through tool calling.
How to give enterprise AI agents real autonomy in regulated domains — using DSLs, Langium validators, and Mozaik tools that embed legal rules into the compiler loop.
Most frameworks build autonomous agents. The next step is autonomous teams: specialized agents that observe, communicate, and collaborate inside Mozaik's event-driven environment.
Three developers found Mozaik, read the code, disagreed with zero of the architectural decisions, and started shipping. This is how v3.11–v3.13 happened in two weeks.
After six years as a software engineer, Eric Evans' Domain-Driven Design transformed how I model Mozaik, core domains, and event-driven AI agents.
In Mozaik, streams are semantic events other agents can react to.
Production logs aren't observability for multi-agent AI. kaleidoskop replays every Mozaik orchestration as a scrubbable hex grid — bus events ripple, agents light up, you find the moment that broke the run in seconds.
A friendly walkthrough: terminal chat with Ink, agent wiring with Mozaik, and shell tools—plus a degit-ready starter repo.
How recent Mozaik releases reframe the framework around reactive agents - with typed event handlers, plain-text messages on the bus, the self vs. external split, and a foundation for collaborative agent swarms like baro.
Mozaik 3.0 evolves from a multi-provider library into a TypeScript framework for non-blocking AI agents - with structured context, explicit context memory, async context item generators, and a shared AgenticEnvironment for multi-agent collaboration.
Most GenAI frameworks ignore context as a first-class primitive. Learn how Mozaik introduces a rich domain model for context engineering in agentic applications.
Why unified LLM abstractions fail and how modeling each LLM separately improves developer experience. A lesson in Domain-Driven Design from building Mozaik.
Showing 13 of 13 stories
You're all caught up.
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts
