Senior Engineer for RAG Support Assistant
Budget / SalaryHourly project
TypeFreelance project
LocationRemote
Posted14 hours ago
I run a small software agency, and I need a senior AI full stack engineer to build an AI customer support assistant for one of our clients. The client is a mid sized online business. They want an assistant that answers customer questions from their own help docs and hands harder cases to a human agent.
Scope of work:
A document ingestion pipeline that takes PDFs, web pages and FAQ files, splits them into chunks, creates embeddings and stores them in a vector database.
A RAG backend API in Python with FastAPI, or in Node.js, that pulls the relevant content and generates answers using the OpenAI or Anthropic API, with source citations.
A chat widget in React that can be embedded on the client’s website.
An admin dashboard in Next.js where the client can upload documents, view chat history, flag wrong answers and see basic usage stats.
Human handoff, where the assistant passes the conversation to a support agent when it can’t answer with confidence.
Deployment on AWS with Docker, plus a short setup and handover document.
Client communication:
You’ll join calls with me and the client to confirm requirements, explain technical tradeoffs and walk them through progress. Spoken English at C1 level or higher is required. Before the project starts, I’d like a short voice or video call through Freelancer’s built in call feature, so please don’t bid if you only want to work through text chat.
Requirements:
Senior level full stack experience. Production experience with LLM APIs, RAG pipelines and vector databases such as Pinecone, pgvector or Weaviate. Strong Python or TypeScript. Experience deploying on AWS with Docker.
Milestones:
Milestone 1 is the ingestion pipeline and RAG API. Milestone 2 is the chat widget and human handoff. Milestone 3 is the admin dashboard, deployment and handover.
In your bid, please include:
A short description of an AI or RAG project you’ve shipped and what your part was, which tech stack you’d pick for this project, and your estimated hours for each milestone.
Skills: Python, FastAPI, Node.js, React.js, Next.js, TypeScript, OpenAI, LangChain, Machine Learning (ML), Artificial Intelligence, Natural Language Processing, AWS, Docker, PostgreSQL
Budget: Hourly, 40 to 50 USD per hour
Scope of work:
A document ingestion pipeline that takes PDFs, web pages and FAQ files, splits them into chunks, creates embeddings and stores them in a vector database.
A RAG backend API in Python with FastAPI, or in Node.js, that pulls the relevant content and generates answers using the OpenAI or Anthropic API, with source citations.
A chat widget in React that can be embedded on the client’s website.
An admin dashboard in Next.js where the client can upload documents, view chat history, flag wrong answers and see basic usage stats.
Human handoff, where the assistant passes the conversation to a support agent when it can’t answer with confidence.
Deployment on AWS with Docker, plus a short setup and handover document.
Client communication:
You’ll join calls with me and the client to confirm requirements, explain technical tradeoffs and walk them through progress. Spoken English at C1 level or higher is required. Before the project starts, I’d like a short voice or video call through Freelancer’s built in call feature, so please don’t bid if you only want to work through text chat.
Requirements:
Senior level full stack experience. Production experience with LLM APIs, RAG pipelines and vector databases such as Pinecone, pgvector or Weaviate. Strong Python or TypeScript. Experience deploying on AWS with Docker.
Milestones:
Milestone 1 is the ingestion pipeline and RAG API. Milestone 2 is the chat widget and human handoff. Milestone 3 is the admin dashboard, deployment and handover.
In your bid, please include:
A short description of an AI or RAG project you’ve shipped and what your part was, which tech stack you’d pick for this project, and your estimated hours for each milestone.
Skills: Python, FastAPI, Node.js, React.js, Next.js, TypeScript, OpenAI, LangChain, Machine Learning (ML), Artificial Intelligence, Natural Language Processing, AWS, Docker, PostgreSQL
Budget: Hourly, 40 to 50 USD per hour
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.