AI Agent & RAG Developer — LangChain, OpenAI & Vector Databases
Muhammad Ali builds AI agents and Retrieval-Augmented Generation (RAG) systems that answer from a company's own documents instead of guessing, and that can take actions through connected tools.
Discuss Your ProjectWhat I Build
- RAG pipelines that detect new documents and update embeddings automatically
- Chatbots and knowledge bases that stay current
- AI agents with tools, memory and multi-step logic
- AI automation in custom Python
Tools I Use
- OpenAI API & Agents SDK
- Claude API
- Google Gemini
- LangChain
- LangGraph
- CrewAI
- AutoGen
- MCP Servers
- Pinecone
- Qdrant
- Weaviate
- Chroma
- Supabase
- Python
AI Agents & RAG Projects
VideoAutomated RAG System with n8n, Google Drive & Vector DB
An n8n workflow that detects new or updated PDFs in Google Drive, creates OpenAI embeddings and keeps a vector database in sync, so chatbots answer from current documents.
- n8n
- Google Drive API
- OpenAI Embeddings
- Vector Database
- RAG
AI Newsletter Scraper & Email Generator (Python + Streamlit)
A Python and Streamlit tool built for a Fiverr client that scrapes products from their website, fills a themed newsletter template and sends it by email in seconds.
- Python
- Streamlit
- Web Scraping
- LLM APIs
- Email Automation
Autonomous Meta Outreach Engine (FB & IG)
A production-ready Python engine that runs human-like cold outreach on Facebook and Instagram, with ChatGPT-personalized intros and a strict 3-step follow-up sequence.
- Python
- Selenium
- OpenAI API
- Google Sheets API
VideoAI Video Localization Dashboard with Voice Dubbing & Lip-Sync
A dashboard for a client in Morocco that turns a YouTube link or uploaded video into a dubbed, lip-synced version in another language, end to end.
- AI Voice Dubbing
- Lip-Sync AI
- Video Translation
- Web Dashboard
Frequently Asked Questions
What is a RAG system?
RAG (Retrieval-Augmented Generation) lets an AI answer from your own documents. Content is turned into embeddings and stored in a vector database, and the most relevant parts are retrieved for each question.
How is a RAG knowledge base kept up to date?
In his n8n pipeline, any new or updated PDF in Google Drive is detected automatically, processed into OpenAI embeddings and stored in the vector database.