Agentic AI Engineer

Senior Associate Agentic AI Engineer

AI Engineering & Intelligent Automation | AI Managed Services

Role Snapshot

Role

Senior Associate Agentic AI Engineer

Experience

5 to 8 years

Work Location

Anywhere in India (Preferably Hyderabad / Bangalore)

Education

Bachelors in Computer Science, Engineering, or related field (Masters or cloud/AI certifications preferred)

Core Skills

Agentic AI Workflows, LLM Orchestration, Python, FastAPI, AWS (Bedrock), Docker & Kubernetes, Redis/ElastiCache

Nice to Have

MCP tools, Langfuse, Evaluation harnesses, AWS certifications, Enterprise AI governance

About the Role

As a Senior Associate – Agentic AI Engineer, you will design, build, and operationalize agentic AI solutions using modern LLM orchestration frameworks and cloud-native architectures. Working alongside senior engineers, architects, and operations teams, you will deliver scalable, secure, and governed AI workflows that move reliably from development into production.

This is a hands-on, engineering-focused role covering agent design, orchestration, evaluation, and release readiness within an enterprise AI managed services environment.

Key Responsibilities

Agentic AI Workflow Development

Design multi-agent systems that coordinate reasoning, tool use, memory, and task execution using LangGraph, CrewAI, AutoGen, and similar frameworks. Implement MCP (Model Context Protocol) tools and custom tool interfaces to extend agent capabilities.

LLM Orchestration & Prompt Engineering

Orchestrate LLM interactions with LangChain across retrieval, tools, memory, and agents. Design, test, and optimize prompts for reliability, performance, and cost — supporting versioning, experimentation, and controlled rollouts.

Backend & API Engineering

Develop Python-based AI services using FastAPI. Expose agent and workflow capabilities via secure, scalable REST APIs, with asynchronous workflows, background tasks, and event-driven processing where appropriate.

Cloud-Native AI Platform Development

Build and deploy AI services on AWS using Bedrock for foundation model access. Integrate IAM, logging, and monitoring to meet enterprise security and compliance requirements, while optimizing for performance, scalability, and cost.

Containerization & Deployment

Package services with Docker and deploy to Kubernetes. Support deployment pipelines that enable consistent builds across dev, test, and production environments in collaboration with platform and operations teams.

State, Memory & Caching

Design and implement agent memory and caching strategies using ElastiCache (Redis), optimizing retrieval, session state, and intermediate results for performance and reliability.

Observability, Guardrails & Evaluation

Implement guardrails for safety, compliance, and reliability (input validation, output constraints, tool-use controls). Instrument workflows with Langfuse for tracing and evaluation, and build evaluation harnesses to validate quality and regression risks before release.

Release, Versioning & Collaboration

Contribute to release planning by validating workflow readiness and evaluation results. Use GitHub for version control, pull requests, and code reviews, and collaborate closely with architects, product owners, and operations teams.

Continuous Improvement & Learning

Stay current with emerging agentic AI frameworks and LLM capabilities. Identify opportunities to improve reliability and developer experience, and contribute reusable components and best practices to shared repositories.

Required Skills & Experience

• Strong Python development experience for AI and backend services.
• Hands-on experience with agentic AI frameworks (LangChain, LangGraph, CrewAI, AutoGen, or similar).
• Experience building REST APIs using FastAPI.
• Working knowledge of AWS cloud services, including AWS Bedrock.
• Experience with Docker and Kubernetes for containerized deployments.
• Familiarity with Redis / ElastiCache for caching or state management.
• Experience with prompt engineering, prompt testing, and optimization.
• Exposure to guardrails, observability, and evaluation for LLM-based systems.
• Proficiency with GitHub workflows and collaborative development practices.