Role Overview :

As an Engineer - Agentic AI, you will build autonomous, multi-step AI agents on AWS that reason, plan, and act across enterprise systems. You will work with Amazon

Bedrock, Amazon Bedrock Agents, and AgentCore to architect solutions where AI agents orchestrate tools, APIs, and knowledge bases to automate complex workflows.

This role sits at the intersection of Generative AI, software engineering, and enterprise integration.

Key Responsibilities :

- Design and implement agentic AI workflows using Amazon Bedrock Agents and

AWS AgentCore - including tool definition, action groups, and multi-step reasoning chains.
- Build retrieval-augmented generation (RAG) pipelines using Amazon Bedrock

Knowledge Bases, Amazon OpenSearch Serverless, and Amazon Aurora pgvector.
- Integrate agents with enterprise APIs and data sources via AWS Lambda, Amazon

API Gateway, and AWS Step Functions.
- Develop prompt engineering frameworks - system prompts, chain-of-thought,

few-shot patterns - optimised for Claude, Titan, Llama, and Mistral models on Amazon Bedrock.
- Implement agentic memory, session management, and context persistence using

Amazon DynamoDB and ElastiCache.
- Configure Amazon Bedrock Guardrails for content filtering, PII redaction, topic

restrictions, and hallucination controls.
- Build and test agent evaluation pipelines using Amazon Bedrock Evaluations to

measure accuracy, groundedness, and tool-selection correctness.
- Deploy and monitor agentic applications using AWS Lambda, Amazon

ECS/Fargate, CloudWatch, and AWS X-Ray for trace-level debugging.
- Contribute to proof-of-concept engagements and client-facing Agentic AI

architectures.

Technical Skills Required :

AWS Agentic AI Services (Primary)

Confidential

Quadrasystems.net

Quadrasystems.net India Private Limited

AWS SBU - Job Descriptions 2026

- Amazon Bedrock - foundation models, Bedrock Agents, Knowledge Bases,

Guardrails, Evaluations
- AWS AgentCore - Runtime, Gateway, Memory, Identity, Policy,



Observability

modules
- AWS Lambda, Amazon API Gateway, AWS Step Functions
- Amazon OpenSearch Serverless, Amazon Aurora pgvector, Amazon DynamoDB

Underlying AI/ML Technologies (Broadly Applicable)

- Large Language Models (LLMs) - Claude, GPT-4, Llama, Mistral - architecture

and behaviour
- Agentic frameworks - LangChain, LangGraph, CrewAI, AutoGen, Strands Agents

(AWS preferred)
- RAG patterns - chunking strategies, embedding models, hybrid search, re-ranking
- Prompt engineering - system prompts, chain-of-thought, ReAct patterns, tool-use

schemas
- MCP (Model Context Protocol) - tool definitions and agent-tool integration
- Python - primary language for agent development (boto3, requests, Pydantic)
- Vector databases - OpenSearch, Pinecone, pgvector, Chroma (conceptual)
- REST APIs, OAuth 2.0, JSON schema for tool definitions

Valuable to Have :

- AWS Certified AI Practitioner or AWS Certified Machine Learning - Specialty
- Exposure to multi-agent orchestration patterns (A2A, hierarchical, swarm)
- Familiarity with fine-tuning workflows on Amazon SageMaker or Bedrock

Qualifications :

- Bachelors degree in Computer Science, Artificial Intelligence, Software

Engineering, or a related field
- 2+ years of software engineering experience; at least 1 year working with LLMs or

Generative AI in a development capacity
- Hands-on experience building or contributing to AI-powered applications or

automation workflows

Why Quadrasystems :

Our Agentic AI practice is among the most active in the AWS partner ecosystem in India.

You will work on enterprise-scale agentic architectures using the latest Amazon Bedrock capabilities - from AgentCore Runtime to multi-agent orchestration - with direct mentorship from senior architects and exposure to cross-industry client outcomes.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.

📌 Engineer - Agentic AI (Bengaluru)
🏢 Quadrasystems.Net
📍 Bengaluru