Hacktoberfest 2025: Our Team’s Open Source Contributions in AI and Technical Debt Management
At OmbuLabs, open source has always been at the core of our values. This year, for Hacktoberfest 2025, our team took the opportunity to give back to the communities that power our daily work, from Ruby developers tackling technical debt to those building the next generation of AI tools. Throughout October, we dedicated focused internal time to open source contributions. Our goal was twofold:
- Strengthen and maintain the tools that power our work at FastRuby.io, helping teams manage technical debt and improve code quality.
- Advance Ruby’s presence in artificial intelligence, contributing to libraries and frameworks that integrate Ruby with modern AI technologies. By aligning Hacktoberfest participation with our mission to give back, we turned this month into an opportunity for growth, learning, and meaningful community impact.
The Projects and Focus Areas
We organized our work around our two brands, FastRuby.io and OmbuLabs.ai , each representing a distinct but complementary area of our expertise. Unlike in past years, we didn’t split into separate smaller teams. Instead, our developers collaborated across projects, sharing ideas and pairing up to merge dozens of pull requests. This approach fostered deeper collaboration and allowed us to make broader contributions across multiple repositories.
FastRuby.io: Tackling Technical Debt
For the first three weeks of October, our attention was centered on our projects that improve code quality, maintainability, and technical debt management. Our team contributed to several open source tools that are widely used in the Ruby community:
- RailsBump.org : A tool to check your Ruby gems for compatibility with all Rails minor versions.
- RubyCritic : A static analysis tool that visualizes code quality metrics.
- Skunk : A gem that gives another metric to help developers identify code that would benefit from refactoring.
- next_rails : A gem that assists with Rails upgrades by identifying compatibility issues.
- reek : A classic Ruby tool for detecting code smells.
- ruby_or_rails : A Discord bot that shares puzzles about different Ruby and Rails methods.
- benchmark-fyi : A project to share benchmark-ips data for all your benchmarks.
OmbuLabs.ai: AI Meets Ruby
In the final week of Hacktoberfest, we turned our focus to the AI side of our work at OmbuLabs.ai. With the growing interest in integrating Ruby applications with large language models (LLMs), our team explored and contributed to tools that bridge the gap between Ruby and AI frameworks. We decided to focus our contributions on the following projects:
- Active Agent : A framework to allow Rails applications to incorporate agents.
- DSPy.rb : A Ruby port of Stanford’s DSPy library for programmatic prompting.
- langchain.rb : A Ruby port of the popular LangChain framework, designed for building powerful applications on top of LLMs.
These projects reflect our commitment to making Ruby a first-class citizen in the world of AI, empowering developers to experiment, prototype, and build intelligent systems without leaving the Ruby ecosystem.
Our Contributions
We were able to contribute a lot this Hacktoberfest, especially in our FastRuby.io projects. Here’s what we were able to ship this October.
RubyCritic
Julio and I focused on removing parser warnings in Ruby 3.4 and above.
- PR #539 : Updated RubyCritic to support the new Prism parser in Ruby 3.4 and above.
Skunk
Henrique , Juan , and Francois contributed to Skunk in multiple ways, including adding Ruby 3.4 support, adding HTML report generation, and refactoring Skunk’s analysis code.
- PR #124 : Added Ruby 3.4 CI compatibility and gem updates.
- PR #123 : Introduced HTML report generation, improved JSON output structure, and added a new configuration class for format selection.
- PR #127 : Refactored the analysis layer to centralize Skunk attributes within RubyCritic’s module collection, simplifying the architecture and improving maintainability.
ruby_or_rails
Fiona , Juan , Francois , Mateus , Julio and Rishi made numerous usability and UI improvements to our community puzzle bot:
- PR #35 : Added the ability to update puzzles directly from the admin table.
- PR #38 : Used
simple_formatto preserve line breaks and improve puzzle readability. - PR #40 : Added validations to the
Puzzlemodel. - PR #34 : Displayed “Sent” dates in a readable format.
- PR #37 : Fixed validation issues when seeding data.
- PR #44 : Improved the edit modal for better text area usability.
- PR #46 : Displayed the total count of puzzles in each table.
Together, these changes made the bot easier to use, maintain, and contribute to, both for our internal team and for the Ruby community at large.
Benchmark.fyi
Ariel upgraded Rails and refactored code to make it adhere to Rails conventions:
- PR #16 : Updated the platform to make Rails 7.1 the current version.
- PR #15 : Refactored models and controllers to follow Rails conventions, including strong parameters, validations, and better JSON handling.
RailsBump
Mateus and Julio upgraded the application and enhanced the UI:
- PR #126 : Upgraded the app to Rails 8.0 and removed Yarn for a cleaner, dependency-free setup.
- PR #125 : Enhanced the compatibility data table, adding richer metadata to the rake task for better tracking.
- PR #128 : Improved the UI for large tables with horizontal scroll support and fixed CI package installation issues.
Dspy.rb
Francois improved the modularity and reduced core gem dependencies.
- PR #178 : Extracted OpenAI adapter to
dspy-openaigem.
Looking Ahead
Hacktoberfest isn’t just about open source contributions; it’s about strengthening the ecosystem we all rely on. This year’s efforts reinforced our belief that investing in open source pays dividends not just for the community, but also for our own growth as developers and teams.
We’re proud of what we accomplished and excited to continue maintaining and improving the tools that make our work possible, whether that’s helping teams upgrade legacy Rails applications or pioneering the next generation of AI integrations in Ruby.