Patchwork – Open-Source AI Workflows for Dev Chores
Hi HN, We are excited to share patchwork - an open-source CLI for dev chore automation that you can use with your LLM of choice. Dev Teams can orchestrate custom workflows (called ‘patchflows’) using a combination of reusable steps and prompt templates to fix vulnerabilities, upgrade breaking dependencies, generate documentation, and more. We built scanning tools in the past, and saw how overwhelmed developers get with their DevSecOps pipelines. LLMs have the potential to help - but there is a need for an 'outer-loop' solution that can be customized to accommodate the processes, priorities,…
What it does
In the maker’s words, at launch
Hi HN, We are excited to share patchwork - an open-source CLI for dev chore automation that you can use with your LLM of choice. Dev Teams can orchestrate custom workflows (called ‘patchflows’) using a combination of reusable steps and prompt templates to fix vulnerabilities, upgrade breaking dependencies, generate documentation, and more. We built scanning tools in the past, and saw how overwhelmed developers get with their DevSecOps pipelines. LLMs have the potential to help - but there is a need for an 'outer-loop' solution that can be customized to accommodate the processes, priorities, and insights of different teams. Even better if the solution can be run locally for security and compliance reasons. Patchwork is built with these needs in mind - it gives devs full control over how and where they want to use LLMs in their SDLC. Patchwork can serve as a universal open framework for your dev chore automation, unlike point (and closed) solutions that focus on specific tasks. Open-sourcing Patchwork was a no-brainer. Agentic workflows fail far too often and need deep context. We believe that the first generation of truly usable AI agents will need users to define and customize their behavior without constraints. And open source is the best way to enable that. Plus, as a team, we have been part of initiatives like LF/OpenSSF and love the community. We are an early, bootstrapped team and would love for you to try patchwork and give us some feedback. Tell us what chores or tasks you'd like to automate, or what must we add before you can adopt something like this!
Does the same job
all alternatives →
- GBGraph-Based Editor for LLM Workflows2024 · github.com · ▲8
Hey HN, We’re excited to share PySpur, an open-source tool that provides a graph-based interface for building, debugging, and evaluating LLM workflows. Why we built this: Before this, we built several LLM-powered applications that collectively served thousands of users. The biggest challenge we faced was ensuring reliability: making sure the workflows were robust enough to handle edge cases and deliver consistent results. In practice, achieving this reliability meant repeatedly: 1. Breaking down complex goals into simpler steps: Composing prompts, tool calls, parsing steps, and branching…
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 · 16d 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
