Senior AI Engineer Generative AI And AWS

LetzBizz; Chennai; Hybrid; INR 2,600,000 - 5,200,000; Full time

Job summary

LetzBizz in Chennai seeks an experienced AI/ML engineer to design, develop and deploy AI applications on AWS, combining generative AI, autonomous agents and machine learning. You will move solutions from experimentation to production, ensuring reliability, security and ongoing operation.

Responsibilities include building agent orchestration, RAG pipelines, secure integrations, prompts and evaluation metrics, plus CI/CD, observability and knowledge transfer across teams.

Qualifications

• 6–8 years of experience in AI/ML, data science or software engineering
• Strong Python programming and production engineering skills
• Hands-on experience with Bedrock, including foundation models, Knowledge Bases, Guardrails
• Experience with GenAI and RAG, embeddings, vector databases and retrieval strategies
• Experience with agent orchestration, planning, memory, tool invocation and human approval workflows
• Working knowledge of supervised learning, feature engineering, model evaluation, deployment, drift detection and monitoring
• Experience integrating solutions using AWS Lambda, Step Functions, OpenSearch, S3 and APIs
• Hands-on experience with Git, automated testing, CI/CD, containers and infrastructure automation
• Ability to work with US-based stakeholders, explaining solution behavior and trade-offs

Responsibilities

• Develop AI agents capable of planning tasks, using tools, maintaining context and coordinating workflows
• Build retrieval-augmented generation (RAG) solutions including embeddings, vector searches and grounded responses
• Connect AI apps to enterprise APIs and AWS services through secure integrations
• Implement prompts, structured responses, conversation state and routing
• Establish automated evaluations for accuracy, groundedness, relevance, safety, latency and cost
• Apply guardrails, data protection, audit logging and human review controls
• Develop conventional ML workflows: feature engineering, training, tuning, deployment and retraining
• Implement testing, CI/CD, infrastructure automation and observability
• Collaborate with architects, data engineers and client stakeholders to deliver testable capabilities
• Support demos, user acceptance testing and production releases

Skills

Python

AWS Bedrock

GenAI

RAG

Embeddings

Vector databases

Agent orchestration

CI/CD

Git

OpenSearch

S3

Lambda

Step Functions

Containers

Tools

Amazon Bedrock

Amazon SageMaker

LangChain

OpenSearch

S3

AWS Lambda

Step Functions

Git

CI/CD

Terraform or AWS CDK

Job description

Job Description:

6–8 Years

Hybrid / Remote

Chennai / Kerala / Tamil Nadu

Hiring Partner: LetzBizz

Client Location: Poland

Location: Chennai / Kerala / Tamil Nadu

Work Mode: Hybrid or Remote

About The Role

We are looking for an experienced engineer to develop and deploy AI applications on AWS, combining generative AI, autonomous agents and machine learning. You will take solutions from experimentation through production, with responsibility for reliability, evaluation, security and ongoing operation.

Key Responsibilities

• Develop AI agents capable of planning tasks, using tools, maintaining context and coordinating workflows, including human approvals where required.
• Build retrieval-augmented generation (RAG) solutions covering document ingestion, chunking, embeddings, metadata, vector and hybrid search, reranking and grounded responses.
• Connect AI applications to enterprise APIs, data platforms and AWS services through secure, permission-controlled integrations.
• Implement prompts, structured responses, conversation state, routing, fallbacks and error recovery.
• Establish automated evaluations for response accuracy, groundedness, retrieval relevance, safety, task completion, latency and cost.
• Apply guardrails, personal-data protection, audit logging and human review controls.
• Develop conventional ML workflows where needed, including feature engineering, training, tuning, deployment, monitoring and retraining.
• Implement testing, CI/CD, infrastructure automation, model and prompt versioning, observability, alerting and operational documentation.
• Work with architects, data engineers, application teams, domain specialists and client stakeholders to define and deliver testable capabilities.
• Support demonstrations, user acceptance testing, production releases, troubleshooting, performance improvements and knowledge transfer.

Required Skills And Experience

• 6–8 years of experience in AI/ML, data science or software engineering, with strong Python programming and production engineering skills.
• Hands-on experience with Amazon Bedrock, including foundation models, Knowledge Bases, Guardrails, model evaluation and agent or tool-calling implementations.
• Strong practical understanding of GenAI and RAG, including embeddings, vector databases, retrieval strategies, reranking and prompt engineering.
• Experience with agent orchestration, planning, memory, tool invocation, human approval workflows and multi-agent coordination.
• Working knowledge of supervised learning, feature engineering, model evaluation, deployment, drift detection and monitoring.
• Experience integrating solutions using AWS Lambda, Step Functions, OpenSearch, S3 and APIs, with secure access to enterprise data.
• Hands-on experience with Git, automated testing, CI/CD, containers and infrastructure or configuration automation.
• Ability to work directly with US-based business and technical stakeholders, explaining solution behaviour, evaluation results, limitations and engineering trade-offs.

Preferred Experience

• Amazon SageMaker for developing and operating ML pipelines.
• Amazon Bedrock AgentCore, Strands Agents, LangGraph, LangChain or comparable agent frameworks.
• OpenSearch vector search or knowledge graphs.
• AI delivery in Life Sciences or Healthcare, particularly research, clinical operations, medical affairs, commercial operations, manufacturing or pharmacovigilance.
• Familiarity with regulated environments and GxP requirements.
• Responsible AI, red teaming, synthetic evaluation data, LLM observability, fine tuning or multimodal applications.
• Terraform or AWS CDK.
• Relevant AWS certifications in AI or machine learning.

What Success Looks Like

Delivering secure, reliable AI applications with measurable quality, effective agent execution, monitored performance and cost, reusable components, and a successful transition from proof of concept to production.