AI Product Engineer – AI Products & Agents

About the role

Oliver Wyman Digital is hiring an AI Product Engineer in Dubai/Abu Dhabi to build the AI products and agentic solutions behind the firm's AI-first transformation work for clients and internally. The engineer owns how an AI product is built, secured and made reliable, from data and retrieval through agent orchestration, integration, user interfaces, deployment, evaluation and ongoing improvement, working in small multidisciplinary teams alongside AI Product Owners, consultants, industry experts and client teams.

Key responsibilities

• Build full-stack AI products: user interfaces, backend services and APIs with appropriate frontend frameworks and UX principles.
• Develop single and multi-step agents with tools, state, memory and human checkpoints, and optimise their instructions, context and tool specifications.
• Design agent workflows with durable state, resumability, failure recovery, safe retries, timeouts and rollback or compensation.
• Build retrieval and knowledge pipelines, prepare structured and unstructured data, and design knowledge stores with data quality, lineage, protection and access controls.
• Deploy AI applications on cloud platforms such as AWS, integrating with enterprise systems via APIs, and support client cloud and on-premise environments.
• Apply secure engineering and least-privilege access, handle untrusted content, and support IT, Risk and Legal approvals.
• Build evaluation and regression-testing practices, monitor quality, latency and cost, and work with client technology or platform teams to support products in production.
• Translate product objectives into technical decisions, explain trade-offs to non-technical stakeholders, and prepare client teams to run the products.
• Use AI coding tools productively, contribute reusable components and standards, and eventually lead workstreams and mentor others.

Required profile

• Degree in computer science, engineering or a related field, or equivalent practical experience.
• Software engineering experience through employment, projects, coursework or open-source contributions; production delivery experience is valued for more senior candidates.
• Ability to build working LLM-powered applications or agents and explain what worked and what failed.
• Willingness to grow across user interfaces, backend services and full-stack delivery.
• Excellent written and verbal communication in English.

Required skills

• Strong coding in Python and/or TypeScript, with debugging, source control and automated testing; code review and CI/CD exposure is an advantage.
• Foundational data skills including SQL.
• Hands-on cloud experience, preferably AWS, including containers, APIs and infrastructure.
• Understanding of cloud security, identity and access management, networking, monitoring and cost optimization.
• Awareness of AI-specific risks such as prompt injection, unsafe tool use, secrets exposure and data leakage.
• Understanding of software testing and scenario-based evaluation of agent outputs.

Nice to have: multi-agent orchestration, vector or graph databases, observability tooling, AWS services such as Bedrock, Lambda, ECS/EKS, S3, RDS, IAM and CloudWatch, infrastructure as code, AI security and governance frameworks, and consulting or client-facing experience. A portfolio of code, demos or shipped products is also a plus.