Building Aktilot: An Open-Source AI Workspace for Engineering Teams

Aktilot is an open-source AI workspace designed for engineering teams. It connects platforms like GitHub, Jira, Confluence, Slack, and other enterprise tools into a single intelligent workspace.

By combining Retrieval-Augmented Generation (RAG), and both cloud and local LLMs, Aktilot enables developers to search organizational knowledge, understand codebases, answer technical questions, and automate engineering workflows—all from one place.

Modern engineering knowledge is scattered across multiple tools.

• GitHub for code
• Jira for tasks
• Confluence for documentation
• Slack for discussions
• Google Drive for design documents

When developers need answers, they often switch between all of these systems.

I started building Aktilot to solve this problem.

What is Aktilot?

Aktilot is an open-source AI workspace that connects engineering tools and allows developers to ask questions in natural language.

Instead of searching across multiple applications, developers can ask questions like:

• Where is this feature implemented?
• Which Jira ticket introduced it?
• Is there documentation for this API?
• Where was this discussed?

Architecture

Aktilot is built around four core components:

Connectors : GitHub, Jira, Confluence, Slack, Google Drive
Knowledge Engine : Collects, indexes, and retrieves information
AI Layer : Supports OpenAI and local LLMs like Ollama
Web UI : A single interface to search and interact with your
engineering knowledge

Current Features

• ✅ GitHub Connector
• ✅ AI Chat Interface
• ✅ OpenAI Support
• ✅ Ollama (Local LLM) Support
• ✅ Extensible Connector Framework

What's Next?

I'm currently working on:

• Jira Connector
• Confluence Connector
• Slack Connector
• Google Drive Connector
• More AI-powered engineering workflows

Why Open Source?

I wanted Aktilot to be self-hostable, extensible, and community-driven. My goal isn't just to build another chatbot—it's to create a workspace where AI can understand an engineering team's knowledge across different systems.

The project is still evolving, and I'd love feedback from the community.

GitHub: https://github.com/your-org/aktilot
website: https://aktilot.com