[Hiring] AI Engineer @Owlis Technology Limited

Role Description

When a client decides to use AI, someone has to build it, connect it to their systems and make it safe. This role works directly with clients, often on site, to build and deploy AI assistants, agents and automations on their data, then hands them over in a state our service desks can support.

Core systems

• Azure OpenAI, Azure AI Foundry and Azure AI Search
• Copilot Studio and Power Platform
• Python and/or C#, Git, Azure DevOps or GitHub
• Microsoft Graph and client system APIs
• HaloPSA for project time, tasks and handover to support

What you will be doing

• Design, build and deploy LLM applications for clients: assistants, retrieval over documents, agents and AI-assisted workflows
• Integrate with client systems through APIs, with scoped permissions
• Choose models and services to fit the client’s needs and budget, with the Head of AI
• Use version control, testing and deployment pipelines for everything we ship
• Work on site or remotely alongside client staff to understand the work and test with real users
• Explain what was built, how to use it and its limits in plain English
• Gather feedback after go-live and improve the build
• Build evaluation sets and monitoring for accuracy, cost and usage
• Apply least privilege, secrets management and client data handling rules
• Write handover documentation and runbooks for our service desks
• Record AI systems in our register with owners and risk ratings
• Contribute connectors, templates and prompt and evaluation sets to a shared library
• Build internal AI tools for our own group companies
• Log time and tasks in HaloPSA

Outcomes and KPIs

• In the first 90 days: several client builds running, with reusable components (connectors, prompt and evaluation sets) in our shared library.
• By 6 months: 10 or more repeatable propositions delivered and rolled out to clients, each with standard scope, testing and handover.
• By 12 months: revenue targets met on the first 10 propositions, with more in flight.

Competencies

• Clear evidence of building and shipping LLM applications to real users, with integrations and testing described
• Explains retrieval or tool design, prompt and model choices, cost control and failure handling from first-hand work, and is honest about mistakes
• Builds integrations and automations with error handling, logging, least privilege and scoped service accounts, and uses vendor features before custom code
• Listens to clients and gets examples, explains plainly what the tool can and can’t do, and agrees a fix and a follow-up
• Designs a clear architecture (sources, indexing, permissions-aware retrieval, model, interface), talks through trade-offs and adapts when requirements change
• Uses security-trimmed retrieval, a tested evaluation set with agreed accuracy, and monitoring with a feedback loop, keeping data in the client’s tenant where possible
• Owns mistakes and fixes them, follows through and manages expectations when things slip

Essential skills and experience

• Has built and shipped applications using large language models, such as assistants, retrieval over documents (RAG) or agents
• Writes production-quality code in Python or C#, with version control and testing
• Has integrated AI or applications with business systems through APIs, for example Microsoft Graph, a CRM or a ticketing tool
• Builds with security and data protection in mind: least privilege, secrets handling and safe use of client data
• Can work directly with client users to understand a problem and explain what was built

Desirable skills and experience

• Experience with Azure OpenAI, Azure AI Foundry, Azure AI Search or Copilot Studio
• Has set up evaluation for AI outputs (test sets, accuracy measures, monitoring)
• Power Platform, Logic Apps or another workflow and integration tool
• Experience in an IT managed services provider or a client-facing engineering role
• Microsoft Azure AI Engineer Associate (AI-102) or similar certification
• Infrastructure as code or CI/CD pipelines in Azure

Development

We expect the role to grow into a senior or lead AI engineer as the practice scales. Certification (for example AI-102) is supported where it fits the work.

Our recruitment process

• A short pre-screening call with our HR / Recruitment team.
• A competency interview with the Head of AI / Operations Director
• A live system design exercise with a client scenario and a technical assessment
• A short final call with an Owlis Group executive, after which we aim to make an offer.

Equal opportunities

The Owlis Group is an equal opportunities employer. We welcome applications from everyone and make recruitment decisions on the basis of skills, experience and potential, regardless of age, disability, gender reassignment, marriage or civil partnership, pregnancy or maternity, race, religion or belief, sex or sexual orientation. If you need any adjustments to take part fully in the recruitment process, please let us know and we will be glad to help.