I will build your rag chatbot, vector database and ai automation system

Hi, I'm Khem I build AI chatbots and automation systems that actually know your business, using Retrieval-Augmented Generation (RAG) and vector databases instead of a generic prompt-only bot that guesses or makes things up.

I connect your docs, APIs, databases, and workflows into an AI assistant that answers from your real data not from training data alone. Whether you need a support chatbot embedded on your site, an internal tool that answers questions from your company's own documents, or an automation pipeline that hooks your AI up to real business logic (not just chat), I can build it end-to-end.

What I work with:

• RAG pipelines chunking, embeddings, retrieval, reranking
• Vector databases Pinecone, pgvector, Weaviate, Chroma
• LLM integration OpenAI API, Claude/Anthropic API, LangChain
• Chatbot UI & embeds React, Next.js, embeddable widgets
• Automation connecting your AI to real actions, not just conversation
• Backend Node.js, PostgreSQL, AWS S3 for document/file handling

Recent work:

• Lexa my own RAG chatbot builder. Teams sync a GitHub repo, upload PDFs, or point it at a help center; it handles chunking, embeddings, and tone/guardrail configuration, then deploys as an script.