About AZ4C
AZ4C is a technology company focused on helping organisations apply cloud, data and AI to real business challenges. We design and deliver practical solutions on the Microsoft ecosystem, combining strong engineering with close client collaboration to move ideas from proof of concept to reliable production systems.
Key Responsibilities
- Build agent features, tools and connectors on Microsoft Agent Framework and Microsoft Foundry from agreed designs.
- Build and maintain data pipelines and models in Microsoft Fabric, Azure SQL and Data Factory that agents and reports depend on.
- Write Azure Functions and APIs that agents call, with error handling, retries and logging.
- Prepare test data, write evaluation cases and run evaluations before each release.
- Add tracing and logging to your code; investigate and fix issues found in testing and production.
- Use AI coding assistants responsibly, reviewing and testing everything they produce.
- Follow security practices: managed identities, Key Vault and no secrets in code.
- Document your work and take part in stand-ups, code reviews and client demos.
Required Skills & Experience
- 1–2 years of hands-on experience in software, data or AI engineering.
- Good Python and SQL; comfortable with APIs, Git and basic testing.
- Has built at least one LLM application: retrieval, tool calling or an agent.
- Working knowledge of Azure services such as Azure OpenAI or Foundry, Functions and Storage.
- Understanding of data pipelines, data modelling and data quality.
- Clear communication and the ability to learn quickly as tools change.
Preferred Qualifications
- Bachelor's degree in Engineering, Computer Science, AI and Data Science, IT or a related field.
- Exposure to Microsoft Fabric, Power BI or Azure Data Factory.
- Exposure to Microsoft Agent Framework, Semantic Kernel, LangGraph or similar agent frameworks.
- Microsoft certifications such as AI-900, DP-900, AI-102 or DP-700.
Success Measures
- Assigned tasks are delivered on time, tested and accepted in code review with little rework.
- Agents and pipelines you build pass evaluation and acceptance checks before release.
- Within six months, you own a feature independently from build to production.