Druids – Build your own software factory

Hi HN!Druids (https://github.com/fulcrumresearch/druids) is an open-source library for structuring and running multi-agent coding workflows. Druids makes it easy to do this by abstracting away all the VM infrastructure, agent provisioning, and communication. You can watch our demo video here (https://www.youtube.com/watch?v=EVJqW-tvSy4) to see what it looks like.At a high level:- Users can write Python programs that define what roles the agents take on and how they interact with each other.- A program is made of events - clear state transitions that the agents or clients can call to modify state. Each event gets exposed as an agent tool.- Druids provisions full VMs so that the agents can run continuously and communicate effectively.We made Druids because we were making lots of internal coding tools using agents and found it annoying to have to rearrange the wiring every time.As we were building Druids, we realized a lot of our internal tools were easier to express as an event-driven architecture – separating deterministic control flow from agent behavior – and this design also made it possible to have many agents work reliably.We had issues with scaling the number of concurrent agents within a run, so we decided to have each program run in an isolated sandbox program runtime, kind of the same way you run a Modal function. Each agent then calls the runtime with an agent token, which checks who can talk to who or send files across VMs, and then applies the tool call.Our early users have found the library useful for:- running many agents to do performance optimization- building custom automated software pipelines for eg code review, pentesting, large-scale migrations, etc...We've heard that the frontier labs have the infrastructure to quickly spin up 100 agents and have them coordinate with each other smoothly in various ways. We're hoping that Druids can be a starting point to make that infrastructure more accessible.

Hacker NewsApr 8ToolDevSource

FL score

out of 100

Verdict

BUILD

high confidence

Competition

competitors found, emerging market, funded players

Trend

No signal yet

An open-source library for building and orchestrating reliable, scalable multi-agent coding workflows by abstracting infrastructure.

The pain

Developers building multi-agent systems struggle with the complexity of provisioning VMs, wiring agents, managing communication, and ensuring reliability and scalability (context degradation, long execution times, difficulty integrating).

The gap

While many AI coding assistants and open-source agent frameworks exist, few provide a dedicated, abstracted layer for VM infrastructure, agent provisioning, and event-driven, sandbox-isolated communication specifically designed for building *custom, reliable multi-agent coding workflows*. The gap is in simplifying the 'factory' infrastructure for developers rather than just offering agents or basic frameworks.

Build angle

Create a managed service or premium extensions for Druids, focusing on specific high-value use cases (e.g., performance optimization, large-scale migrations) for teams that need to scale and reliably coordinate numerous autonomous agents without managing underlying infrastructure.

Strengths

Addresses a real, specific, and growing pain in multi-agent orchestration.
Strong creator-problem fit, born from personal experience.
Clear technical differentiation with infrastructure abstraction and event-driven design.

Questions about this idea?

FlyBot reads the scoring and gives you a second opinion on “Druids – Build your own software factory”.

Open FlyBot

Fly Labs Method

Is the pain real, is there a gap, is it the right time, can one person build it.

Problem clarity: 25
Solution gap: 14
Willingness to pay: 19
Buildability: 11

Druids addresses a real and specific pain in multi-agent orchestration for developers, with a clear technical differentiation in infrastructure abstraction. However, the market is crowded and the build complexity for a solo builder is high, especially for a foundational library.

Value Equation

Dream outcome and how likely it feels, against the time and effort it costs.

Good market viability and value proposition in a rapidly growing space, but needs a clear monetization strategy and stronger differentiation to thrive.

One-Person Business

Curiosity pull, identity fit, and a path from free value to paid for a solo creator.

Strong problem clarity and creator fit for a complex but impactful developer tool, but needs a clearer monetization strategy and careful management of complexity.

Viral Frameworks

Hook strength, shareability, and how cheaply it can be tested.

A clear value proposition for a specific developer audience, but needs a defined business model and further validation on the commercial side.

Builder Lens

Evidence the problem exists, timing, defensibility, and a model that fits on a napkin.

Addresses a clear, growing need in the AI agent space with a well-defined initial product, poised for future relevance.

Why this verdict

The idea has strong fundamentals: a clear problem, a competent builder, and a growing market. However, the 'crowded' competition level and the inherent complexity of the build (especially for a solo builder trying to create a foundational library) indicate that a pure 'build' recommendation is too risky without further validation. The critical next step is to validate a specific commercial offering and ensure there's enough willingness to pay to sustain a solo operation, given the existing open-source competition and the high cost of alternative proprietary solutions.

Five lenses, one composite. How scoring works

The angle

This weekend

Identify the most painful, specific use case for Druids that a small group of users would pay for as a managed service.
Validate demand and pricing for a commercial offering (e.g., hosted Druids environment or premium features) with current and potential users.
Focus marketing efforts on developer communities experiencing the specific pain points Druids solves for multi-agent workflows, potentially showcasing how it solves issues like Devin AI's context degradation or integration challenges.

Who is already there, emerging market

Devin AI (Cognition Labs)FundedVisit
An autonomous AI software engineer capable of planning, executing, debugging, and iterating across an entire codebase with minimal human input.
Pricing: $500/month (reported)
CursorVisit
An AI-native code editor built on top of VS Code that provides AI pair programming features, including inline editing, multi-file AI editing sessions, and codebase-aware chat.
Pricing: Not explicitly found, but referred to as potentially having paid plans.
Windsurf (formerly Codeium)FundedVisit
An AI-powered IDE and AI coding assistant with an agentic AI system called Cascade that executes multi-step coding tasks and provides deep codebase understanding.
Pricing: Free when bringing your own API keys.
OpenHands (formerly OpenDevin)FundedVisit
An open-source autonomous agent framework designed to execute software tasks with minimal supervision, including modifying codebases, running shell commands, browsing the web, and interacting with APIs.
Pricing: Open-source (MIT licensed), users pay for LLM API costs.
SWE-agentVisit
An open-source AI software engineer that performs well on the SWE-bench benchmark for evaluating developer skill, capable of correcting a significant percentage of submitted bugs.
Pricing: Open-source
AiderVisit
An open-source command-line AI coding assistant that integrates with Git repositories, excelling at multi-file editing through natural language commands.
Pricing: Free (open-source), users pay for LLM API costs.
MetaGPTVisit
A multi-agent framework that aims to replace most of Devin's capabilities, including data interpretation, solving math problems, and web crawling.
Pricing: Open-source
Devika AI (Stition AI)Visit
An open-source AI software engineer tool designed to rival Devin AI, capable of understanding high-level human instructions, breaking them down into steps, researching, and writing code.
Pricing: Open-source
GitHub CopilotFundedVisit
An AI-powered development assistant that integrates directly into IDEs, providing inline code suggestions, conversational interfaces, and refactoring capabilities.
Pricing: Team ($19/user/month), Business ($39/user/month). (Based on typical Copilot pricing, specific enterprise pricing for 'AI software factory' use case not found).
JetBrains JunieFundedVisit
An AI-powered coding agent integrated into JetBrains IDEs that assists with code completion, testing, and inspections, adapting to workflows and collaborating on complex tasks.
Pricing: Not explicitly found, but offered within JetBrains ecosystem which has various pricing tiers.
Qodo (formerly CodiumAI)FundedVisit
A multi-agent code integrity platform that enhances code quality with AI-powered tools for writing, testing, and reviewing code, integrating with IDEs and Git platforms.
Pricing: Not explicitly found, but offers flexible deployment options and enterprise features. (Free tier for PR-Agent functionality is open-source.)
Amp CodeVisit
An AI coding assistant available as a VS Code extension and CLI that enables developers to delegate complex coding tasks, refactors, reviews, and explorations to intelligent agents.
Pricing: Not explicitly found.

Funded players

What they charge

The pricing landscape for AI coding tools varies significantly. Open-source solutions like OpenHands, SWE-agent, Devika AI, and Aider are generally free, with users only incurring costs for the underlying LLM API usage. Proprietary tools like Devin AI are reported to be expensive (e.g., $500/month). Other commercial offerings like GitHub Copilot have subscription models (e.g., $19-$39/user/month). Many solutions offer free tiers or trials, while enterprise-grade solutions often feature custom pricing, advanced security, and on-premise deployment options.

Druids – Build your own software factory

Excessive CAPTCHA challenges degrading GitHub login UX for enterprise users

GTA modders tediously manually edit assembly files for modpacks

BreezePDF – Free, in-browser PDF editor