An open-source Elixir application for sharing daily learnings with a 200-word limit per post.
Tilex is a Today I Learned (TIL) platform built in Elixir that enables teams to catalog and share knowledge as it accumulates day-to-day. It features concise posts limited to 200 words to encourage focused, digestible insights and supports controlled collaboration through flexible authentication. Originally developed internally at Hashrocket, it was open-sourced to provide insight into development processes and allow community experimentation.
Development teams and organizations looking to implement a structured, internal knowledge-sharing platform for capturing daily technical learnings. It's particularly suited for teams using Elixir who want a self-hosted solution with AI integration capabilities.
Developers choose Tilex for its simplicity, practicality, and focus on capturing fleeting insights in a lightweight manner. Its unique selling points include AI-powered post creation via MCP servers, flexible authentication with domain restrictions, and being production-ready with comprehensive deployment guides.
Today I Learned
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
The 200-word post limit ensures insights are focused and digestible, directly aligning with the README's philosophy of capturing fleeting knowledge in a structured way.
Integrates MCP servers for AI-assisted post creation, allowing users to generate content with tools like Claude, as detailed in the AI Tooling section with setup snippets and API key generation.
Uses Ueberauth with Google OAuth, supporting domain restrictions and allowlists for controlled access, which is configurable via environment variables per the Authentication setup.
Provides comprehensive setup with Docker, Heroku deployment guides, and CI integration, making it self-hosting ready with clear instructions from the Installation section.
Authentication is limited to Google OAuth via Ueberauth, with no built-in support for other providers like GitHub or SAML, which the README admits requires specific Google configurations.
Built entirely in Elixir and Phoenix, requiring Erlang, Node, and PostgreSQL setup, which adds complexity for teams unfamiliar with this stack and limits portability.
AI integration relies on MCP servers and external tooling, which the README notes requires additional setup like generating API keys, potentially creating a barrier for non-technical users.