Senior Applied AI Engineer

Design and operate production agentic AI services for AI-first customer support at Azure scale. In this role at Microsoft (ACES CØDE), you will help improve diagnostics and documentation using LLM orchestration, RAG, evaluations, and production engineering, turning experimentation into measurable customer outcomes. This is an onsite position in Redmond, WA.

What you’ll do

• Analyze support-case trends, customer issue patterns, and authorized technical signals at scale to uncover recurring failure patterns and gaps in Azure diagnostics and documentation.
• Build scalable agentic AI solutions for diagnostics and documentation, using LLM orchestration, RAG pipelines, agent workflows, tool integrations, and memory and state management, including human-in-the-loop experiences.
• Write, review, test, and maintain production-quality C# and Python while building reusable components, reference implementations, and repeatable engineering patterns.
• Convert support scenarios and product requirements into technical designs, then take solutions through implementation, testing, deployment, monitoring, and continuous improvement.
• Develop evaluation frameworks with rubrics, golden datasets, judge models, and offline and online evaluation, using measurable quality thresholds to move systems from experimentation to production operations.
• Ensure services meet requirements for correctness, reliability, availability, latency, scalability, maintainability, cost, and safety.
• Apply privacy protections and agent-security controls, including prompt-injection defenses, least-privilege tool access, identity and authorization controls, data boundaries, approval gates, auditability, and fail-safe behavior.
• Ground agent responses and actions in trusted, current signals by integrating with Azure data sources, APIs, and cloud-native services.
• Deliver production readiness with CI/CD, automated testing, feature flags, staged rollouts, tracing, logging, metrics, and alerting.
• Diagnose production issues, participate in incident response, and drive corrective improvements.
• Evaluate models, frameworks, prompts, retrieval approaches, and architectural options using experiments and production telemetry, balancing quality with reliability, latency, cost, maintainability, and operational complexity.
• Instrument solutions and use data to improve automation quality, case-volume reduction, resolution time, engineering efficiency, and customer satisfaction.
• Lead design reviews, resolve technical challenges, coordinate implementation across contributors, and align partner teams around shared interfaces and delivery goals.
• Support other engineers through code reviews, technical guidance, reusable documentation, and practical coaching in applied AI and production software engineering.
• Partner with product managers, researchers, platform engineers, Azure service teams, and support-domain experts to build durable solutions and shape future platform capabilities.

Required qualifications

• Bachelor’s Degree in Computer Science or related technical field AND 4+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python (or equivalent experience).
• Ability to meet Microsoft, customer and/or government security screening requirements for this role.
• Microsoft Cloud Background Check: required to pass upon hire/transfer and every two years thereafter.

Preferred qualifications

• Experience building or integrating with LLM-based systems (prompt engineering, RAG, or agent frameworks).
• Familiarity with Azure services such as Azure OpenAI, Azure AI Foundry, Azure Functions, Cosmos DB, or similar cloud-native technologies.
• Experience with CI/CD pipelines, automated testing, and production observability.
• Experience with Azure AI Foundry, Azure OpenAI, Microsoft Agent Framework, or comparable enterprise AI platforms.
• Experience designing AI evaluation systems using rubrics, golden datasets, judge models, simulation, adversarial testing, and offline or online evaluation.
• Experience with RAG systems including indexing, retrieval, ranking, grounding, citation, knowledge freshness, and access control.
• Experience implementing agent-security patterns such as prompt-injection defense, identity propagation, least-privilege tool access, data-boundary enforcement, approval gates, and auditable actions.
• Experience building reliable distributed services with strong standards for scalability, resiliency, performance, observability, and cost efficiency.
• Experience with infrastructure as code, container orchestration, feature management, and staged production deployment.
• Demonstrated ability to lead technical execution, influence design decisions, and mentor engineers without formal people-management authority.
• Experience collaborating with product, platform, research, and partner engineering teams on complex co-development initiatives.
• Strong business judgment connecting engineering investments to customer and operational outcomes.
• Background in customer support engineering, supportability, diagnostics, or customer-experience platforms is a plus.

Compensation

USD 119,800 - 261,000 per year.

Technologies you may work with

• C, C++, C#, Java, JavaScript, Python
• LLM orchestration, RAG pipelines, agent workflows, Azure OpenAI, Azure AI Foundry, Azure Functions, Cosmos DB
• CI/CD, Microsoft Agent Framework, feature management, container orchestration, infrastructure as code
• Prompt engineering, rubrics, golden datasets, judge models, simulations, offline evaluation, online evaluation
• Azure data sources, tracing, logging, metrics, alerting