Agentic AI Architect

Dentsu Global Services Private Limited; Karnataka; Hybrid; INR 3,500,000 - 5,200,000; Full time

Job summary

Merkle, part of Dentsu, seeks an experienced AI engineering leader to act as the technical anchor for multi-agent systems. You will drive architecture decisions, set engineering standards, and mentor a team of agentic engineers while remaining hands-on with code.

You will lead LLM orchestration, design robust AI workflows, and work closely with onshore teams to align strategy with execution. This role is permanent and based in India with Bengaluru as a key location.

Qualifications

• 12–15 years of software engineering experience with at least 4 years in AI/ML engineering.
• Deep expertise in Python and production-grade software architecture.
• Hands-on experience building and deploying LLM-powered applications at scale.
• Strong knowledge of agent orchestration patterns (LangChain, LangGraph, AutoGen, or custom frameworks).
• Experience with cloud infrastructure (Azure preferred) including container orchestration and serverless patterns.
• Track record of mentoring engineers and raising team engineering standards.
• Experience with MLOps: model serving, monitoring, versioning, and cost management.

Responsibilities

• Lead technical architecture for multi-agent systems, LLM orchestration, and AI automation workflows.
• Set up and maintain the agentic engineering stack: frameworks, CI/CD, testing, monitoring, and deployment patterns.
• Own the technical design for AI orchestration use cases across client implementations.
• Build and maintain reusable agent components, tool-use libraries, and prompt templates.
• Lead Genie space setup and configuration for client data exploration.
• Mentor and technically lead 3 Senior Agentic Engineers, conducting code reviews and architecture discussions.
• Evaluate LLM providers (model selection, cost optimization, latency tradeoffs) and recommend choices.
• Bridge the gap between Data Science model outputs and agentic deployment, ensuring models are served efficiently.
• Collaborate with the onshore Senior AI Engineer on shared architecture decisions and technical standards.
• Contribute to technical documentation, runbooks, and operational playbooks for the agentic platform.

Skills

Python

Software architecture

LLM deployment

LangChain

LangGraph

AutoGen

Azure

Tools

Databricks

Unity Catalog

Genie

Job description

Job Description

Location: Bengaluru, Mumbai, Pune, Gurugram Experience - 12-15 years Role Summary Serve as the technical anchor for Dentsu's agentic AI engineering team in DGS. You will lead architecture decisions for multi-agent systems, set engineering standards, and mentor a growing team of agentic engineers. This is a hands-on leadership role: you write code, review code, and set the technical direction while the Senior Director handles the VP-level strategy and stakeholder management.

Key Responsibilities

• Lead technical architecture for multi-agent systems, LLM orchestration, and AI automation workflows
• Set up and maintain the agentic engineering stack: frameworks, CI/CD, testing, monitoring, and deployment patterns
• Own the technical design for AI orchestration use cases across client implementations
• Build and maintain reusable agent components, tool-use libraries, and prompt templates
• Lead Genie space setup and configuration for client data exploration
• Mentor and technically lead 3 Senior Agentic Engineers, conducting code reviews and architecture discussions
• Evaluate LLM providers (model selection, cost optimization, latency tradeoffs) and recommend choices
• Bridge the gap between Data Science model outputs and agentic deployment, ensuring models are served efficiently
• Collaborate with the onshore Senior AI Engineer on shared architecture decisions and technical standards
• Contribute to technical documentation, runbooks, and operational playbooks for the agentic platform

Required Qualifications

• 12 - 15 years of software engineering experience with at least 4 years in AI/ML engineering
• Deep expertise in Python and production-grade software architecture
• Hands-on experience building and deploying LLM-powered applications at scale
• Strong knowledge of agent orchestration patterns (LangChain, LangGraph, AutoGen, or custom frameworks)
• Experience with cloud infrastructure (Azure preferred) including container orchestration and serverless patterns
• Track record of mentoring engineers and raising team engineering standards
• Experience with MLOps: model serving, monitoring, versioning, and cost management
• Strong system design skills with experience in distributed systems and API architectureExcellent written communication for technical documentation and cross-timezone collaboration

Preferred Qualifications

• Experience with Databricks, Unity Catalog, and/or Genie
• Background in media technology, adtech, or marketing analytics platforms
• Experience with RAG architectures, vector databases, and semantic search at scale
• Contributions to open-source AI/ML projects
• Experience leading technical teams in a DGS/offshore delivery model

Location: DGS India - Bengaluru - Manyata N1 Block

Brand: Merkle

Time Type: Full time

Contract Type: Permanent

About dentsu

For over 120 years, innovation has been a core tenet of our offering - exploring new ways to reach, engage and nurture relationships with audiences. Together we drive a multiplier effect for clients at a global scale, through the development of Integrated Growth Solutions that are underpinned by our promise to clients: innovating to impact. Be a force for good. Sustainability is a vital part of our business and an important area of focus for our clients. We're leading the way - helping to build a more sustainable planet. Dream loud. In this moment of transformation, we need our people to be fearless, embracing change and ambiguity, driven by the love for their work and excitement for the future. Team without limits. We create opportunities for connection and collaboration between our colleagues and clients, building a sense of belonging and having some fun along the way. Find out more about us

Who we are

Our Social Impact

Our work