Maarga Systems - AI Practice Leader

Description :

Role Overview :

Title : AI Practice Leader

Level : Senior Individual Contributor / Technical Lead

Experience : 8-12 years

Location : Chennai (hybrid; client travel as needed)

Maarga is building its Agentic AI practice. This is a foundational hire the person who defines how Maarga delivers AI, not just what it delivers. You will be the technical anchor on client engagements, the internal standard-setter for architecture and delivery quality, and a co-creator of Maarga's AI point of view.

This is not a manager role and not a pure researcher role. It is a builder-leader : someone who architects, delivers, mentors, and engages clients with equal confidence.

Why This Role Exists :

AI specifically Agentic AI is moving faster than any single enterprise can absorb. Our clients are running AI programs at scale and are struggling with two things : too many disorganized experiments, and not enough depth to build the right architecture. Maarga's edge is the ability to bring cross-industry, bleeding-edge expertise with the humility to learn alongside the client.

We need someone who can walk into a room of data scientists, enterprise architects, and senior program managers and hold their own on all three fronts.

Technical Architecture :

- Design multi-agent AI systems with sophisticated control flows, inter-agent communication, and state management

- Architect context management strategies : knowledge graphs, GraphRAG, dynamic context injection, chunking approaches

- Determine when to use LLM-based reasoning versus deterministic logic and articulate that reasoning to clients

- Evaluate and recommend agentic frameworks (LangGraph, AutoGen, Semantic Kernel, custom solutions) based on client context

- Define integration patterns for enterprise systems, data sources, and existing automation pipelines

- Establish security and governance models for autonomous agent operations (agent identity, access control, audit trails)

Delivery Leadership :

- Anchor client engagements as the senior technical voice from discovery to production deployment

- Create reusable frameworks, blueprints, and reference architectures that accelerate future Maarga engagements

- Define architectural patterns and guard against anti-patterns (e.g., monolithic super-agents, over-reliance on LLMs)

- Own technical quality across agentic solutions from context pipeline design to evaluation frameworks

- Guide and upskill junior Maarga team members on agentic patterns and production discipline.

Practice Building :

- Contribute to Maarga's AI point of view : what we believe, what we recommend, and why

- Develop assessment tools, proposal templates, and engagement playbooks for AI/agentic work

- Stay current on the frontier (new models, frameworks, evaluation techniques) and translate it into practical client guidance

- Participate in business development : support proposals, client conversations, and technical scoping

Required Skills :

- 5+ years building production distributed systems, microservices, or orchestration platforms

- 2+ years hands-on experience with LLM APIs (Anthropic Claude, OpenAI, Azure OpenAI, or equivalent)

- Strong Python programming; comfortable with async, event-driven, and workflow orchestration patterns

- Production experience with at least one major cloud platform (Azure, AWS, or GCP)

- Solid understanding of RAG patterns, embeddings, vector databases, and semantic retrieval

- Experience designing systems that handle uncertainty and non-deterministic outputs

- Consulting or client-facing experience comfortable presenting, questioning, and co-designing with senior stakeholders

- Strong written and verbal communication can explain complex architecture simply

Desired Skills (Not All Required) :

- Hands-on with agentic frameworks : LangGraph, LangChain Agents, AutoGen, CrewAI, Semantic Kernel

- Knowledge graph design and reasoning (Neo4j, Microsoft GraphRAG, LLM-generated graphs)

- Background in MLOps or LLMOps : model monitoring, drift detection, evaluation harnesses

- Microsoft ecosystem depth : Azure AI services, Copilot Studio, M365 Copilot, Power Platform

- Understanding of enterprise AI governance agent identity, access scoping, audit

- Exposure to SAFe or scaled agile delivery methodologies

- Experience in FMCG, retail, supply chain, or media domains (a plus, not a requirement)