Retrieval-Augmented Generation (RAG) Engineer

Yantran; Chennai District; On-site; INR 1,200,000 - 2,400,000; Full time

Job summary

Yantran is seeking a RAG Engineer in Chennai to design and develop retrieval-augmented generation pipelines. You will implement document ingestion, chunking, embeddings, and vector/hybrid search to connect LLMs with enterprise knowledge bases.

The role involves building production-grade APIs, evaluating performance, and optimizing retrieval accuracy across cloud-based platforms using LangChain, LlamaIndex, and modern vector databases.

Qualifications

• Experience designing and building RAG pipelines for AI apps.
• Ability to build document ingestion and processing pipelines.
• Experience with chunking and embedding strategies.

Responsibilities

• Design and develop RAG pipelines.
• Build document ingestion and processing pipelines.
• Implement chunking and embedding strategies.
• Configure vector and hybrid search.
• Integrate LLMs with enterprise knowledge bases.
• Optimize retrieval accuracy.
• Implement reranking and metadata filtering.
• Evaluate RAG performance.
• Develop production APIs and services.

Skills

RAG

LLMs

Python

Embeddings

Vector databases

Pinecone

Weaviate

Qdrant

FAISS

Chroma

Elasticsearch

OpenSearch

LangChain

LlamaIndex

REST APIs

Cloud platforms

Tools

Pinecone

Weaviate

Qdrant

FAISS

Chroma

Elasticsearch/OpenSearch

Job description

Seeking a RAG Engineer to build AI applications that combine enterprise knowledge retrieval with Large Language Models.

Responsibilities:

• Design and develop RAG pipelines.
• Build document ingestion and processing pipelines.
• Implement chunking and embedding strategies.
• Configure vector and hybrid search.
• Integrate LLMs with enterprise knowledge bases.
• Optimize retrieval accuracy.
• Implement reranking and metadata filtering.
• Evaluate RAG performance.
• Develop production APIs and services.

Required Skills:

• RAG.
• LLMs.
• Python.
• Embeddings.
• Vector databases.
• Pinecone/Weaviate/Qdrant/FAISS/Chroma or equivalent.
• Elasticsearch/OpenSearch preferred.
• LangChain/LlamaIndex.
• REST APIs.
• Cloud platforms.