Lead AI Engineer (Agentic Systems)

S&P Global Energy seeks a Lead AI Engineer (Agentic Systems) to architect and build production-grade autonomous workflows that reason, plan, collaborate and execute complex tasks. The role combines software, data and machine learning engineering and is listed in Gurgaon, Ahmedabad and Hyderabad, India.

Responsibilities

• Lead hands-on development of stateful, multi-agent systems in Python using orchestration frameworks such as LangGraph, CrewAI or AutoGen. Design agent-to-agent communication, message passing, persistent memory and interruptible control flows for long-running tasks. Implement Model Context Protocol interfaces connecting agents to data and operational tools.
• Manage model routing and fallbacks, context windows and inference costs across proprietary and open-source models. Containerize workloads with Docker, deploy on Kubernetes and use cloud-native infrastructure such as AWS AgentCore.
• Build high-throughput Databricks or Python ETL pipelines that provide structured operational context. Optimize retrieval and vector stores for data with seconds-to-minutes latency; work with data engineering and AI teams on schemas, pipeline requirements and data availability.
• Implement tracing and observability for reasoning steps, token usage and latency. Build human-in-the-loop controls, break-glass mechanisms and guardrails; automate evaluations of agent accuracy, hallucinations and drift. Set architecture and code standards, review processes and the agentic roadmap, mentor engineers and prototype emerging approaches including graph-based RAG.

Required qualifications

• 7+ years of total technical experience in software engineering, data engineering or machine learning, including 2+ years building and deploying LLM applications or agentic systems in production.
• Advanced Python; experience architecting AI storage with vector, relational or NoSQL databases and Databricks or Snowflake; expertise deploying and scaling applications with cloud services, Kubernetes and Docker. Familiarity with LLM orchestration, RAG, embeddings and context management, plus data science concepts, prompting, CI/CD, API design, asynchronous programming and reliability.
• Bachelor’s degree in computer science, engineering, mathematics or a related technical field.

Preferred qualifications

• Master’s degree or PhD in a relevant field; 5+ years of hands-on NLP experience; knowledge graphs, graph databases or GraphML; specific agentic tooling and A2A, swarm or multimodal architectures; experience with operational real-time processing in sectors such as FinTech, energy or logistics.

The posting mentions health coverage, time off, learning resources, retirement planning and family-related benefits, but provides no salary amount.