AI Chatbots & RAG Knowledge AssistantsTran Tien Van
I build AI chatbots and knowledge assistants that answer from your own documents, help center and data, and show the source behind each answer.
What you get
A RAG assistant over your docs, PDFs, website or database (pgvector, Pinecone or FAISS)
Built with LangChain / LangGraph and OpenAI or Claude, with guardrails and a hand-off to a person
Deployed where your users are: web chat, Slack, WhatsApp or inside your app
An answer-quality test set, so accuracy is measured instead of guessed
Hosting on AWS, GCP or Azure, plus a short handover
Proof
Graspr (built for a global client): RAG with pgvector, LangChain and Claude raised answer accuracy from about 80% to 95%+
Alice AI (built for a global client): RAG over Pinecone and FAISS inside an assistant with 70+ tool integrations
Yōjin (built for a global client): ChromaDB-backed RAG over a Neo4j threat-and-compliance knowledge graph
Timeline
About 2 weeks for a first assistant on one knowledge base and one channel. If it also needs to take actions in your tools, see my AI Agent Development service.
Starting at$1,500
Duration2 weeks
Tags
FastAPI
LangChain
OpenAI
PostgreSQL
Python
Service provided by
Tran Tien Van proHo Chi Minh City, Vietnam
- 3
- Followers
AI Chatbots & RAG Knowledge AssistantsTran Tien Van
I build AI chatbots and knowledge assistants that answer from your own documents, help center and data, and show the source behind each answer.
What you get
A RAG assistant over your docs, PDFs, website or database (pgvector, Pinecone or FAISS)
Built with LangChain / LangGraph and OpenAI or Claude, with guardrails and a hand-off to a person
Deployed where your users are: web chat, Slack, WhatsApp or inside your app
An answer-quality test set, so accuracy is measured instead of guessed
Hosting on AWS, GCP or Azure, plus a short handover
Proof
Graspr (built for a global client): RAG with pgvector, LangChain and Claude raised answer accuracy from about 80% to 95%+
Alice AI (built for a global client): RAG over Pinecone and FAISS inside an assistant with 70+ tool integrations
Yōjin (built for a global client): ChromaDB-backed RAG over a Neo4j threat-and-compliance knowledge graph
Timeline
About 2 weeks for a first assistant on one knowledge base and one channel. If it also needs to take actions in your tools, see my AI Agent Development service.
$1,500