TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: Jido – Run 10k agents at 25KB each (Elixir)

13 pointsby mikehostetler5 months ago
Hi HN!<p>I&#x27;m Mike Hostetler and I built Jido, an Agent SDK in Elixir that lets you run thousands of agents without heavy infrastructure.<p>Repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;agentjido&#x2F;jido">https:&#x2F;&#x2F;github.com&#x2F;agentjido&#x2F;jido</a> Getting Started: <a href="https:&#x2F;&#x2F;hexdocs.pm&#x2F;jido&#x2F;getting-started.html" rel="nofollow">https:&#x2F;&#x2F;hexdocs.pm&#x2F;jido&#x2F;getting-started.html</a><p>Why another framework?<p>After using several popular Agent frameworks and platforms, I had two key challenges:<p>- Running multiple agents required process-heavy infrastructure like Docker or K8s. Running 50,000 agents in parallel was costly and diminished the benefits of agentic programming.<p>- Today&#x27;s agents require too much human intervention when building workflows. Why couldn&#x27;t agents manage their own WDLC (Workflow Design Life Cycle)? This felt like a major missing piece.<p>Agentic frameworks were written for humans. LLMs working with this code were constantly working around human work-style assumptions. So, I wrote a framework specifically for LLMs to code and operate their own agentic flows.<p>Elixir was a natural choice because of it&#x27;s functional nature, rock-solid concurrency primitives and &quot;let-it-crash&quot; philosophy with dynamic error compensation. Hot code reloading was a bonus.<p>Agents in Jido use 25Kb of memory at rest and can easily serialize then hibernate for long-lived access. Agents possess the APIs to dynamically start and manage their own sub-agents or any other Elixir process utilizing Elixir&#x27;s OTP architecture.<p>Jido Actions are functional primitives that Agents can dynamically orchestrate into workflows. Generated code can either run in a separate process in the current VM or in another BEAM VM that&#x27;s linked and hardened before introduction into the Agent VM.<p>I&#x27;m excited to help enable a world where thousands of agents work seamlessly on behalf of their human operators.<p>Thanks!

4 comments

travisgriggs5 months ago
I used GPT (irony) to (possibly) get to the point of the name:<p>“… • In Japanese, 自動 (Jidō) means “automatic” or “self-moving,” often seen in terms like “jidōsha” (automobile). …”
评论 #42570267 未加载
评论 #42551605 未加载
sbrother5 months ago
This looks great; I&#x27;ve been working on something similar but I&#x27;ll drop it since this is much further along. Do you have a list of issues&#x2F;feature requests you&#x27;d like contributions on?
评论 #42559954 未加载
jittoshi5 months ago
Can you elaborate on the scaling potential here, given your choice of stack?
评论 #42553134 未加载
lizzas5 months ago
Agent as in AI or agent as in actor framework? Or both?
评论 #42552667 未加载