Responsibilities
- Use coding agents to prototype AI solutions quickly against real client requirements, so the team can validate value before committing to a full build
- Build, deploy, and maintain LLM-powered applications and agents that align with enterprise architecture principles and standards, with evaluation, monitoring, and token tracking in place from day one
- Set up MCP connectors and permissions against client systems, taking real care over data handling, access control, and what an agent is and is not allowed to do
- Build production solutions that meet client security requirements and European data protection obligations, including GDPR. This is not optional and not an afterthought
- Collaborate with consultants, business analysts, and architects to break business requirements down into solutions that are feasible, maintainable, and safe
- Create reusable AI assets and accelerators for the practice, including skills, plugins, and MCP servers that make future engagements faster
- Support Data & AI teams with best practices around AI-assisted development, documentation, and automation, and coach less experienced colleagues on agentic build patterns and security standards
- Document deployment, upgrade paths, and maintenance so SQLI and our clients can own what you build after handover
Requirements
- Turn business requirements into working AI solutions using coding agents and frontier models
- Build agentic workflows and LLM-powered applications on top of frontier models
- Wire AI solutions into real enterprise systems and make them safe to run
- Own the build and make real engineering decisions, not just execute someone else’s spec
- Proactively spot better ways to build and be trusted to deliver without close supervision
- Ensure solutions meet client security requirements and European data protection obligations, including GDPR
