Lead AI Platform Engineer

Role Summary / Mission

Design and build AI-Enhanced Software Engineering Platform to accelerate software development and improve developer effectiveness, code quality, and engineering consistency. This is a hands-on platform design and implementation role, not an AI-model development or data-science role. The resource will prioritize pragmatic, high-value MVPs and develop reusable, modular, plug-and-play capabilities that integrate across evolving engineering toolchains.

Key Responsibilities

• Assess the current implementation, define target architecture, transition roadmap, and appropriately scoped MVPs based on value, risk, urgency, and implementation effort.
• Prepare architecture, detailed design, security, data-flow, integration, deployment, and operational-support documentation.
• Present designs and obtain alignment from Information Security, Architecture Review Board, AI Centre of Excellence, Cloud Engineering, Privacy, Risk, and other stakeholders.
• Design and implement modular platform capabilities for governed LLM access, context engineering, enterprise knowledge retrieval, workflow orchestration, security, evaluation, governance, and observability.
• Develop and productionize prioritized engineering productivity use cases, initially integrating with GitHub Copilot, Azure DevOps repositories, work items, pipelines, service hooks, and enterprise knowledge sources.
• Design tool-agnostic interfaces supporting the planned transition to GitLab for repositories, CI/CD and security scanning, and Jira and Confluence for requirements and project documentation.
• Pair with internal engineers to create reference implementations, standards, architecture decisions, runbooks, and knowledge transfer

Required Experience & Skills

• Enterprise AI platform engineering: Hands-on experience designing and productionizing secure, reusable enterprise AI platforms using managed foundation-model services, rather than developing or training AI models.
• GitHub Copilot and governed model access: Experience with GitHub Copilot enterprise capabilities, model selection, licensing and usage controls, and Microsoft-managed access to approved LLMs. Ability to design platform integrations that comply with approved model-access patterns and avoid unmanaged direct connections to public AI providers.
• Context engineering: Strong knowledge of prompt and context design, token management, structured outputs, tool calling, workflow state, grounding, citations, provenance, and prompt and model versioning.
• RAG and vector retrieval: Hands-on experience with embeddings, chunking, metadata design, semantic and hybrid search, retrieval evaluation, and technologies such as Azure AI Search, PostgreSQL with pgvector, or equivalent.
• Toolchain integration and portability: Strong experience with APIs, webhooks, abstraction layers, Azure DevOps, GitLab, Jira, Confluence, CI/CD, repositories, branch and pull-request controls, SAST, DAST, and software-composition analysis; able to design replaceable, loosely coupled components.
• Knowledge architecture: Experience modelling enterprise knowledge, taxonomies, entity relationships, metadata, and provenance using knowledge-graph or graph technologies where justified by the use case.
• AI gateway and model access: Experience implementing governed access to multiple approved LLMs, including model routing, provider abstraction, authentication, policy enforcement, retries, rate limits, token and cost monitoring, model allowlisting, safety controls, and telemetry.
• Security and observability: Strong experience with Microsoft Entra ID, managed identities, Azure Key Vault, least-privilege access, sensitive-data controls, prompt-injection defences, content safety, secure tool execution, audit logging, and OpenTelemetry or Application Insights.