ControlFlow – open-source AI workflows
Hi HN! I’m excited to share ControlFlow, our new open-source framework for building agentic workflows. ControlFlow is built around a core opinion that LLMs perform really well on small, well-defined tasks and run off the rails otherwise. I know that may seem obvious, but the key insight is that if you compose enough of these small tasks into a structured workflow, you can recover the kind of complex behaviors we associate with autonomous AIs, without sacrificing control or observability at each step. It ends up feeling a lot like writing a traditional software workflow. With ControlFlow you:…
In plain words
ControlFlow is an open-source framework for building AI workflows that breaks down complex tasks into small, well-defined steps. It lets developers define structured tasks, assign specialized agents to complete them, and compose these tasks into executable workflows. Built on the principle that language models perform best on specific, bounded problems, ControlFlow aims to provide the complex autonomous behaviors of AI systems while maintaining control and visibility at each step, resembling traditional software workflow development.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi HN! I’m excited to share ControlFlow, our new open-source framework for building agentic workflows. ControlFlow is built around a core opinion that LLMs perform really well on small, well-defined tasks and run off the rails otherwise. I know that may seem obvious, but the key insight is that if you compose enough of these small tasks into a structured workflow, you can recover the kind of complex behaviors we associate with autonomous AIs, without sacrificing control or observability at each step. It ends up feeling a lot like writing a traditional software workflow. With ControlFlow you: - define structured tasks for your agents to achieve, - assign one or more specialized agents to each task, - then compose the tasks into a workflow for execution. We've been working toward this for a while, applying insights from our work on Prefect (our workflow engine) and Marvin (our LLM framework). ControlFlow is our effort to bridge the gap between all those impressive AI demos and what we see B2B users actually trying to accomplish in production. More often than not, their goal is to sanely delegate a few steps of a workflow to an AI, not deploy a suite of end-to-end autonomous while loops. This is ControlFlow's initial release and we’d love your feedback. We hope it's a refreshing take on making LLMs feel like software again! You can check out the repo here: https://github.com/prefecthq/controlflow/
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, June 2024
the whole month →
La Growth Machine▲1,216Create personalized, multi-channel conversations at scale
AI · 2024 · lagrowthmachine.com
PyjamaHR▲1,132Hiring on autopilot. The AI applicant tracking system (ATS).
Work · 2024 · pyjamahr.com