AI Engineer (GenAI/LLMOps)

The AI Engineer will join the Group Functions AI team, focusing on designing and shipping production-grade AI services that power Global Risk, Compliance, Legal, Procurement, and emerging GenAI solutions across the enterprise! The ideal candidate should be an advanced programmer with a strong grasp of Data Science, Data Engineering, Application Development, Software Engineering, and DevOps/LLMOps who has a mix of software craftsmanship, cloud-native engineering, and machine-learning know-how to turn data-science prototypes into reliable, business-critical products!

Position Responsibilities:

• Collaborate with Data Scientists and Data Engineers to build modular, reusable ML & GenAI pipelines (feature stores, model registries, automated retraining).
• Engineer and optimize models for performance, cost, and scalability; run A/B or challenger tests and track data/model drift.
• Own production deployment: containerize models (Docker), orchestrate with Kubernetes/AKS, and automate CI/CD with GitHub Actions or Azure DevOps.
• Stand up production observability for agentic services using OpenTelemetry-based tracing, metrics, and logs, enabling end-to-end visibility across model calls, tool calls, retrieval steps, latency, and errors.
• Build and maintain FastAPI-based and RESTful APIs, with production-grade practices (versioning, validation, auth, error handling, and performance).
• Design monitoring around agent run lifecycle concepts (threads/runs/steps), ensuring every agent execution can be replayed and diagnosed by walking the ordered sequence of operations and nested calls.
• Establish timely/config/workflow versioning and release process rigor: traceable changes, reproducible runs, and safe rollbacks for prompts, agent graphs, retrieval configs, and model settings.
• Guide solution architecture with Cloud & Security teams, network policies, secrets management.
• Implement and productionize GenAI patterns (prompt/context orchestration, RAG, guardrails) on Azure AI Foundry, Databricks, and internal LLMs.
• Champion engineering guidelines and mentor peers on clean code, test automation, and MLOps/LLMOps standards.
• Stay ahead of advances in ML, GenAI, and DevSecOps; introduce fit-for-purpose tools and frameworks to the team.

Required Qualifications:

• 5+ years delivering high-availability software in Python
• Proven experience operationalizing ML/GenAI models in cloud environments (Azure preferred) using Docker, Kubernetes/AKS, MLflow or Azure ML.
• Hands-on with CI/CD pipelines, unit/integration tests, code-quality scanning, and blue-green or canary releases.
• Experience operating LLM/agentic workloads: tracing agent runs (threads/steps/tool calls), debugging multi-step workflows, and controlling cost/latency at scale.
• Experience building and operating data-driven AI/ML applications using Azure Databricks for compute and orchestration, Azure Data Lake for scalable storage, and Cosmos DB for low-latency operational data and application state.
• Experience deploying and/or fine-tuning LLMs and working with vector databases (Azure AI Search, Pinecone).
• Experience with LLM observability and evaluation frameworks (e.g., LangSmith, Langfuse,…) to support production tracing, prompt/version tracking, and offline/online evaluation
• Demonstrated ability to balance urgency with delivering high-quality, well-documented solutions.

Preferred Qualifications:

• Bachelor’s degree in Computer Science, Software Engineering, Data Science, Mathematics, or a related technical field, or equivalent experience in the industry.
• Master’s or PhD degree in a relevant field is an asset and may substitute for part of the required experience.

When you join our team:

• We’ll empower you to learn and grow the career you want.
• We’ll recognize and support you in a flexible environment where well-being and inclusion are more than just words.
• As part of our distributed Group Function AI team, you’ll shape the next generation of AI that drives Manulife’s global success.

#LI-Hybrid