AI Architect, IT Digital Transformation

Position Summary:

Agentic AI Architect

You will architect LLM-powered agents, define robust testing strategies, and drive innovation across agentic automation platforms such as Microsoft Copilot Studio, AWS Agents, and LangGraph

Job Functions and Responsibilities:

AI Architecture & Agent Design

• Architect intelligent agents using LLMs/AWS Bedrock, orchestration frameworks (LangChain, LangGraph), and tool integrations (APIs, databases).


• Enable autonomous task planning, decision-making, and long-running process management.


• Implement memory, context management, prompt engineering, and feedback loops.


• Establish observability, logging, and feedback loops for continuous agent improvement.

Testing & Validation

• Develop comprehensive testing frameworks for agentic systems including unit, integration, regression, and safety tests.


• Design simulation environments to evaluate ethical guardrails, hallucination risks, and performance boundaries.


• Collaborate with QA, risk, compliance, and product teams to align agent behavior with legal and user expectations.

Platform Strategy & Innovation

• Evaluate and prototype agentic frameworks (e.g., CrewAI, AutoGen, Google ADK, etc).


• Lead integration of R&D into production architectures.


• Define architectural blueprints and contribute to internal and external thought leadership.


• Stay current with advancements in cognitive architectures, multi-agent systems, and AI safety

In addition to technical expertise, an Agentic AI Architect must possess a strong set of soft skills to effectively lead the design and deployment of intelligent agent systems. These include:

• Systems Thinking to design holistic, interconnected agent ecosystems.


• Strategic Communication to articulate complex AI concepts to both technical and non-technical stakeholders.


• Collaboration & Cross-Functional Leadership to align AI initiatives across product, engineering, and compliance teams.


• Creative Problem Solving to innovate around the limitations of current LLMs and agentic frameworks.


• Adaptability & Learning Agility to stay ahead in a rapidly evolving AI landscape.


• Empathy & User-Centric Thinking to ensure agents are designed with human needs and usability in mind.


• Decision-Making Under Uncertainty to guide agent behavior in dynamic or ambiguous environments.


• Documentation & Knowledge Sharing to promote reproducibility and team-wide understanding of agentic systems.


• Ethical Reasoning to ensure responsible AI development and deployment.


• Influence Without Authority to drive adoption and alignment without direct control over all stakeholders.

Key Result Areas:

• 1. Agentic System Design & Architecture

  
  
  
  
  • Design scalable, modular, and secure multi-agent architectures using agentic frameworks like AWS Bedrock, LangGraph etc.
  
  
  • Define agent roles, workflows, and interaction protocols aligned with business objectives.
  
  



• 2. LLM Integration & Orchestration

  
  
  
  
  • Integrate large language models (LLMs) into agentic systems with memory, tools, and reasoning capabilities.
  
  
  • Optimize prompt engineering, context management, and tool usage for performance and cost-efficiency.
  
  



• 3. Cross-Functional Collaboration

  
  
  
  
  • Partner with product, engineering, data science, and compliance teams to ensure agentic systems meet functional and regulatory requirements.
  
  
  • Translate business needs into agentic workflows and technical specifications.
  
  



• 4. Innovation & Experimentation

  
  
  
  
  • Lead rapid prototyping and evaluation of new agentic frameworks, tools, and cognitive models.
  
  
  • Benchmark agent performance and iterate on design for continuous improvement.
  
  



• 5. Governance, Safety & Ethics

  
  
  
  
  • Implement safety guardrails, observability, and human-in-the-loop mechanisms.
  
  
  • Ensure compliance with ethical AI principles, data privacy, and organizational policies.
  
  



• 6. Documentation & Knowledge Sharing

  
  
  
  
  • Maintain clear documentation of agent designs, SOPs, and architectural decisions.
  
  
  • Conduct internal workshops or demos to scale agentic AI literacy across teams.
  
  



• 7. Business Impact & Value Delivery

  
  
  
  
  • Measure and report on the ROI of agentic systems in terms of automation, efficiency, and innovation.
  
  
  • Identify new opportunities for agentic AI to drive digital transformation.
  
  

Qualifications:

• Bachelor s or Master s in AI/ML, or Data Science.


• At least 5+ years of relevant experience in AI/ML solutions development.


• At least 2+ years of relevant experience in AI Agentic automation solutions development.


• Proficiency in Python, LangChain, LangGraph, AWS Bedrock, Hugging Face, similar LLM frameworks


• Experience with agent architecture, vector stores, tool calling, and memory management.


• Familiarity with MLOps, model evaluation metrics, and safety techniques for generative AI.


• Experience with cloud platforms (AWS, Azure, GCP), good to have development experience in AWS environments and preferably CI/CD pipelines

WORK SCHEDULE OR TRAVEL REQUIREMENTS

2- 11 PM IST, Mon Fri. No travel.