After 18 months of building, rewriting, and rethinking, Jido 2.0 has shipped. It’s available on Hex now.
Jido started as a bot platform called BotHive in 2024. Then the AI wave hit and everything changed. I was already using Elixir and decided to make a bet: the BEAM is the best runtime for agent systems.
TypeScript agent frameworks felt like toys. Single-threaded event loops trying to juggle concurrent agents with promises and prayer. Python agents did a little better, but after a long time they couldn’t stay up. The BEAM was built for exactly this kind of work.
After working with agents on the BEAM for 18 months, that bet certainly looks to be paying off.
From 1.0 to 2.0
Jido 1.0 was released last March, but admittedly was overengineered. I was still learning OTP in depth, and it showed. I added abstractions that didn’t make sense in practice. This created too much friction to do basic things that other agent frameworks made easy out of the box.
The feedback was clear and consistent. People wanted to build agents, not fight the framework. I took all of that feedback and addressed it in 2.0. Simpler APIs. Less ceremony. BEAM-first from the ground up.
Here’s what we shipped:
A strong, durable agent core
The foundation of Jido 2.0 is a pure functional agent architecture. Agents are data. A struct with state, actions, and tools that can be run inside a GenServer. That’s it.
... continue reading