Added a document intelligence system powered by RAG.
You can upload PDFs, TXT, Markdown, CSV, and JSON files and chat with them using AI.
Documents are chunked, embedded, and queried using vector retrieval before sending context to the LLM.
Document chat is configurable from system settings:
• provider selection (Ollama, Groq, OpenAI, Gemini, HF)
• model override
• top-k retrieval
• temperature control
Next step is integrating document querying directly into workflow automation.
GitHub:
<a href="https://github.com/vmDeshpande/ai-agent-automation" target="_blank" rel="noopener nofollow">https://github.com/vmDeshpande/ai-agent-automation</a>

The official Open Source daily.dev Squad. 

Open Source

Version 0.5.0 of an open-source AI Agent Automation platform introduces a document intelligence system powered by RAG. Users can upload PDFs, TXT, Markdown, CSV, and JSON files and interact with them via AI chat. Documents are chunked, embedded, and retrieved using vector search before context is passed to an LLM. The document chat feature is configurable via system settings, supporting multiple providers (Ollama, Groq, OpenAI, Gemini, HuggingFace), model overrides, top-k retrieval, and temperature control. The next planned step is integrating document querying into workflow automation.

Released v0.5.0 of my AI Agent Automation platform today.