Agentic AI Engineer — Microsoft Azure / Copilot

Vurke is building a global technology-delivery network that supports Microsoft partners, systems integrators, consulting firms, and enterprise clients by providing scalable implementation capacity for complex projects. Unlike traditional recruitment agencies, Vurke’s specialists embed directly into delivery teams on international client engagements. The company focuses on a pre-vetted talent network specializing in Microsoft AI, Copilot, Power Platform, Azure AI, data, and integration projects across North America, the UK, and Europe. This role involves developing production-ready AI agents and multi-agent workflows using Microsoft Foundry, Copilot Studio, and related Microsoft ecosystem tools. You will contribute to current and upcoming client implementations, working within Vurke delivery teams, alongside Microsoft partners, or directly on client projects under agreed delivery models.

Key Responsibilities
Design and develop AI agents that utilize tools, APIs, enterprise data, and workflow actions to automate business processes. Implement agent orchestration using Microsoft Foundry, Copilot Studio, Semantic Kernel, Microsoft Agent Framework, or similar Microsoft-aligned platforms. Build both single-agent and multi-agent workflows with clear role separation, state management, tool schemas, and approval controls. Apply Model Context Protocol (MCP) or comparable tool-integration patterns where relevant. Integrate AI agents with enterprise APIs, Microsoft Graph, Dataverse, SharePoint, Teams, Power Automate, and external systems. Design memory, context, grounding, retrieval, function calling, and structured output patterns to enhance agent capabilities. Implement security features including authentication, authorization, secrets management, data isolation, and human-in-the-loop controls. Develop evaluation suites to measure task success, safety, reliability, grounding, and regression. Add tracing, monitoring, observability, and failure-handling capabilities to support production-grade agents. Collaborate closely with architects, integration engineers, security teams, and business analysts to ensure successful delivery.

Required Qualifications
A minimum of 4 years of professional software engineering experience is required, with at least 2 years focused on generative AI or large language model (LLM) development, preferably involving production deployments. Strong programming skills in Python and/or C# are essential. Candidates should have practical experience with Microsoft Foundry, Azure AI, Azure OpenAI, and/or Copilot Studio. Familiarity with tool calling, retrieval-augmented generation (RAG), agent orchestration, APIs, and structured outputs is necessary. A solid understanding of LLM limitations, prompt injection risks, data leakage, and safe agent design principles is expected. Proficiency in Git, testing methodologies, and API development best practices is also required.

Preferred Qualifications and Additional Information
Experience with multi-agent systems and Model Context Protocol (MCP) is highly desirable. Knowledge of Semantic Kernel, Microsoft Agent Framework, or related orchestration frameworks will be advantageous. Familiarity with Azure AI Search, Microsoft Graph, Power Platform, and Dataverse is preferred. Relevant Microsoft certifications such as AB-620 AI Agent Builder Associate, AI-103 Azure AI Apps and Agents Developer Associate, or AB-100 are valued, especially for senior roles. Candidates should demonstrate a track record of independently delivered projects, showcasing technical depth and client-facing communication skills in English. The ability to work autonomously within distributed teams and align with North American, UK, or European client time zones is important. This engagement is fully remote and contract/project-based, with variable durations depending on client needs. While candidates may be globally located, availability for required time-zone overlaps is essential. Joining the Vurke talent network does not guarantee project assignments; deployment depends on client demand, technical fit, availability, and commercial terms. The recruitment process may include technical assessments and client interviews.

Applicants are encouraged to submit a detailed CV highlighting projects they have personally delivered, technologies used, responsibilities, and measurable outcomes. Hands-on production experience is prioritized over certifications alone, though relevant Microsoft credentials are considered positively.