Agentic AI Developer (AWS)

Cognizant; India; Hybrid; INR 1,800,000 - 3,200,000; Full time

Job summary

Cognizant in Chennai is seeking a Senior Developer to build Agentic AI and GenAI applications on AWS. You will design and implement RAG pipelines, AI agents, and multi-step workflows that integrate with enterprise systems, data stores, and APIs to deliver scalable AI solutions across business units.

You should have strong Python skills and hands-on AWS experience (Bedrock, Lambda, SageMaker, OpenSearch), plus familiarity with LangGraph, AutoGen, or similar frameworks.

Qualifications

• Strong Python skills with hands-on AWS experience (Bedrock, Lambda, SageMaker, OpenSearch).
• Familiarity with LangGraph, AutoGen, or similar frameworks.
• Understanding of RAG, embeddings, and vector search.

Responsibilities

• Build agentic applications using Amazon Bedrock and foundation models.
• Develop AI agents with tool use, memory, and reasoning capabilities.
• Create RAG pipelines and vector search integrations.
• Test, troubleshoot, and support production AI deployments.

Skills

Python

AWS

GenAI concepts

Problem solving

Tools

LangGraph

AutoGen

Bedrock

Lambda

SageMaker

OpenSearch

Job description

Senior Developer Agentic AI (AWS)

Experience: 5 to 8 years (1 to 3 years in GenAI/LLM/RAG)

Location: Chennai

We're looking for a Senior Developer to build Agentic AI and GenAI applications on AWS. You'll develop RAG pipelines, AI agents, and multi-step workflows using frameworks like LangGraph and AutoGen, integrating them with enterprise systems and APIs.

You'll do:

• Build agentic applications using Amazon Bedrock and foundation models
• Develop AI agents with tool use, memory, and reasoning capabilities
• Create RAG pipelines and vector search integrations
• Test, troubleshoot, and support production AI deployments

You should have:

• Strong Python skills and hands-on AWS experience (Bedrock, Lambda, SageMaker, OpenSearch)
• Familiarity with LangGraph, AutoGen, or similar frameworks
• Understanding of RAG, embeddings, and vector search