Lead AI Engineer

Job Title- Lead AI Engineer

Position type- Full Time

Work Location- Bangalore, Noida, Gurgaon

Working style- Hybrid

Shift Time- 12PM to 9PM

People Manager role: No

Required education and certifications critical for the role- BTech in computer scienceor related quantitative field, or equivalent experience.

Required years of experience -

10+ years of work experience in relevant field

AON IS IN THE BUSINESS OF BETTER DECISIONS

At Aon, we shape decisions for the better to protect and enrich the lives of people around the world.

As an organization, we are committed to our purpose as one firm, united through trust as one inclusive, diverse team and we are passionate about helping our colleagues and clients succeed.

GENERAL DESCRIPTION OF ROLE:

We are looking for a Lead AI Engineer to lead a small, high-caliber AI engineering team within our Human Capital Solutions team, embedded within the HCA platform engineering group. This is a hands-on leadership role: you will drive delivery of production-grade AI agents and LLM-powered features that enhance how clients and internal teams interact with Aon's compensation and workforce analytics platform. Working closely with our AI Research and Product Team to adopt established patterns and infrastructure, you will take those patterns and deliver real agentic capability - from natural language interfaces over client data to AI-assisted compensation benchmarking workflows.

The ideal candidate pairs strong Python and backend API development skills with genuine hands-on experience building agentic AI systems and has the technical leadership maturity to give direction to experienced engineers, and own delivery end to end. Given how new this space is, we value deep, demonstrable hands-on capability and sound engineering judgement - whether from professional work or significant personal projects - over years in a formal AI title.

JOB RESPONSIBILITIES :

• Own end-to-end delivery for the team: planning, prioritisation, and predictable execution, so the team makes tangible, visible progress.
• Oversee engineering standards, patterns, and best practices, aligning with the architectural direction set by the AI Research and Product Team.
• Design and build production-grade AI agents using agentic frameworks (initially LangGraph/LangChain, with other frameworks adopted as the space evolves, including exploration of Claude Agent SDK in sandbox environments), integrating with existing platform APIs to deliver intelligent, context-aware features.
• Develop and maintain Python-based backend services (such as FastAPI) that expose agent capabilities via REST and server-sent event (SSE) streaming endpoints.
• Implement secure agent pipelines that correctly propagate user identity and entitlements, ensuring agent behaviour is scoped to individual client subscriptions.
• Apply a strong understanding of AI-specific security concerns - including prompt injection, adversarial inputs, agent behaviour risks, and reputational exposure - to design agents that are robust and trustworthy in a production enterprise environment.
• Integrate AI agents with Azure cloud services and internal platform APIs, following established architectural patterns set by the AI Research and Product Team.
• Instrument agent runs with structured logging, tracing, and observability tooling to support debugging, performance monitoring, and regression prevention.
• Collaborate closely with the HCA platform team in an agile delivery environment - helping steer and contributing to sprint planning, reviews, and ceremonies.
• Work alongside the AI Research and Product Team during onboarding and early delivery, absorbing patterns and progressively taking on more independent ownership of agent design decisions.
• Proactively communicate blockers, unknowns, and design questions - particularly when working at the frontier of rapidly evolving AI tooling.
• Stay current with the fast-moving AI engineering landscape evaluate new frameworks, models, and patterns and bring relevant developments back to the team.

SKILLS/COMPETENCIES REQUIRED :

• 8+ years of overall software development experience, including at least 4 years in Python and backend API development, together with a track record of technical leadership - leading delivery, mentoring engineers, or owning significant systems.
• Substantial, demonstrable hands-on experience building agentic or LLM-powered systems. Given how new this space is, we value depth of capability over years in a formal AI title.
• Proven ability to give technical direction to strong, experienced engineers and to run delivery autonomously, making sound decisions independently and knowing when to align with the wider team.
• Solid understanding of agentic AI concepts and patterns:
• Tool use, state management, and multi-step reasoning
• Checkpointing, human-in-the-loop flows, and agent interrupts
• Multi-agent architectures and orchestration patterns
• Hands-on experience with agentic or LLM frameworks (LangGraph, LangChain, CrewAI, AutoGen, LlamaIndex, Claude Agent SDK, or similar) formal professional experience is a plus but demonstrable hands-on capability - including personal projects and sandbox work - is equally valued.
• Strong Python development skills:
• Async Python and asyncio patterns
• Building and consuming REST APIs and SSE streams
• Python backend frameworks such as FastAPI
• Experience working with Azure cloud services familiarity with deploying and operating cloud-hosted AI workloads on Azure.
• Good understanding of authentication and authorisation patterns (JWT, OAuth2, OIDC) and how to propagate user identity and entitlements securely through an agentic pipeline.
• Awareness of AI-specific security risks including prompt injection, adversarial inputs, and agent behaviour vulnerabilities - and how to design against them.
• Experience with observability practices in backend systems: structured logging, distributed tracing, and monitoring familiarity with AI-specific evaluation and tracing concepts (e.g. MLflow, LangSmith, or similar) is a strong advantage.
• Experience with containerisation (Docker) for packaging and deploying backend services.
• Comfortable using AI coding tools (such as GitHub Copilot, Claude, or Cursor) as part of a day-to-day development workflow.
• Comfortable leading delivery in an agile environment and setting the pace for iterative, dependable execution.
• Excellent communication and stakeholder-management skills proactive in surfacing blockers, unknowns, and design questions across teams and time zones.
• Genuine curiosity and drive for self-directed learning in a fast-moving technical space.

Preferred / Good-to-Have

• Hands-on experience with LangGraph and/or LangChain in a production or near-production context.
• Experience with AWS cloud services (ECS, RDS, ElastiCache, Secrets Manager, S3).
• Familiarity with PostgreSQL and async database access patterns (psycopg3, SQLAlchemy, or similar).
• Experience with agent evaluation frameworks and experiment tracking (MLflow, Databricks Agents, or equivalent).
• Exposure to CI/CD pipelines for AI systems - automated testing, evaluation gates, and deployment of model-backed services.
• Familiarity with Microsoft SQL Server and T-SQL - useful when working alongside the existing HCA platform team.
• Exposure to .NET/C# - helpful for understanding the broader HCA platform codebase these agents integrate with.
• Prior experience building AI features within enterprise SaaS platforms or data-heavy products.
• Relevant certifications (e.g. Microsoft Certified: Azure AI Engineer Associate, AWS Certified Machine Learning - Specialty).