AI Engineer (RAG & Agentic Systems)

DigiRecruitx; Pune District; On-site; INR 800,000 - 1,200,000; Full time

Job summary

DigiRecruitx in Pune, India seeks an hands-on AI Engineer to lead a greenfield RAG system for healthcare credentialing workflows, collaborating with a MERN stack team.

You will design embedding pipelines, vector indexing, and deployment on AWS Bedrock, with REST APIs via FastAPI/Flask and Docker on AWS.

Candidates with 2–4 years of experience, strong Python and LangChain/LangGraph experience are encouraged to apply.

Qualifications

• Strong Python development experience with hands-on RAG/LLM projects.
• Proficient with LangChain and LangGraph for agentic pipelines.
• AWS Bedrock, S3, EC2; IAM/VPC familiarity is a plus.
• Experience with vector databases such as pgvector, Pinecone, or Weaviate.
• Experience building AI backend services and REST APIs using FastAPI or Flask.
• Docker/containerized deployment experience; working with SQL databases; embeddings and semantic search knowledge.

Responsibilities

• Design, develop, and maintain RAG systems, including embedding pipelines and vector indexing.
• Build and maintain knowledge-ingestion pipelines (URL scraping, PDF/DOCX processing, chunking).
• Integrate and optimize LLMs via AWS Bedrock; manage prompts and latency/costs.
• Implement agentic workflows using LangGraph for multi-step reasoning and task orchestration.
• Expose REST APIs to connect AI backend with React-based interfaces and chat apps.
• Implement AI monitoring for logging, hallucination detection, and performance tracking.
• Collaborate with MERN stack team on frontend chat integration and knowledge interfaces.
• Maintain production-grade code and Docker deployments on AWS.

Skills

Python

LangChain

LangGraph

AWS Bedrock

Vector databases

REST APIs

Docker

SQL

Embeddings

Tools

LangChain

LangGraph

Docker

FastAPI

Flask

S3

EC2

Job description

About the Role

We are looking for a hands-on AI Engineer to lead the development of an Agentic RAG (Retrieval-Augmented Generation) system. This is a greenfield AI initiative focused on building an intelligent knowledge assistant for healthcare credentialing workflows, with a roadmap from knowledge Q&A through to automated agentic actions.

You will work closely with the AI Project Lead and collaborate with an existing MERN stack development team to build reliable, production-grade AI capabilities.

Key Responsibilities

• Design, develop, and maintain Retrieval-Augmented Generation (RAG) systems, including embedding pipelines, vector indexing, and retrieval optimization using LangChain and LangGraph.
• Build and maintain knowledge-ingestion pipelines covering URL scraping, PDF/DOCX processing, chunking, embedding generation, and vector database indexing.
• Integrate and optimize Large Language Models through AWS Bedrock, including model selection, prompt engineering, and cost/latency optimization.
• Implement agentic workflows using LangGraph for multi-step reasoning, tool orchestration, and autonomous task handling.
• Build and expose REST APIs using FastAPI or Flask to connect the AI backend with React-based interfaces and chat applications.
• Implement AI monitoring frameworks covering logging, hallucination detection, response-quality evaluation, and performance tracking.
• Collaborate with the MERN stack team on frontend chat integration and knowledge-management interfaces.
• Maintain clean, well-documented, production-grade code and containerized deployments using Docker on AWS.

Required Skills & Experience

• Strong Python skills; Python is the primary development language for this role.
• Hands-on experience with LangChain and/or LangGraph for RAG or agentic pipelines.
• AWS experience, particularly with Bedrock, S3, and EC2; familiarity with IAM and VPC is an advantage.
• Practical experience with at least one vector database such as pgvector, Pinecone, or Weaviate.
• Experience building AI backend services and REST APIs using FastAPI or Flask.
• Hands-on experience with Docker and containerized application deployment.
• Comfortable working with SQL databases and relational data.
• Good understanding of embeddings, semantic search, and chunking strategies for RAG systems.

Good to Have

• React or basic JavaScript knowledge for frontend coordination.
• Experience with Terraform or other infrastructure-as-code tools.
• Familiarity with AI governance, hallucination detection, or response-evaluation frameworks.
• Exposure to healthcare or other compliance-heavy domains.
• Personal or open-source RAG/LLM projects on GitHub are highly valued.

What We Are Looking For

Demonstrated hands-on capability matters more than certifications alone. Candidates who have independently built RAG or LLM-based projects, including personal or open-source work, are encouraged to share GitHub repositories or project demonstrations during the interview process.

CTC: ₹8–12 LPA

Experience: 2–4 Years