Staff Software Engineer, AI (up to £200k)

Dex; United Kingdom; On-site; GBP 120,000 - 180,000; Full time

Job summary

Dex partners with an innovative tax-tech company to build a global AI-powered platform. They are hiring a Staff AI Engineer to shape how LLMs and agentic systems are applied across the product, from prototyping to production, with end-to-end ownership.

You will define the AI strategy and roadmap, design backend services for AI workflows, refine prompt engineering for complex tax scenarios, and build robust evaluation pipelines to ensure safety and reliability in a regulated domain.

Qualifications

• 5+ years building customer-facing products with end-to-end ownership in a high-quality software environment.
• Deep, hands-on experience with LLMs, prompt engineering, and agent frameworks, including production experience with RAG architectures.
• Proven ability to design and implement evaluation pipelines for AI models.

Responsibilities

• Define the AI strategy and roadmap across the product.
• Design and build backend services for AI workflows, from prototypes to production-ready systems.
• Develop and optimise prompt engineering strategies for complex tax compliance scenarios.
• Build robust evaluation pipelines to ensure model accuracy, safety, and reliability in a regulated domain.
• Architect and implement internal and customer-facing agent platforms.

Skills

LLMs

Prompt engineering

Agent frameworks

RAG architectures

Job description

This role is with one of Dex's trusted partner companies. We work closely with their teams to truly understand their culture, goals, and what they're looking for, so we can match you with the right opportunity and give you context about the role before you commit to a process.

The role

This company builds a global tax automation platform, solving genuinely complex regulatory problems for large digital businesses. They're backed by leading investors and have significant ARR, with a clear mandate to apply AI to a domain rich in structured data and multi-jurisdiction edge cases.

This isn't a research role disconnected from production. You'll be the first Staff AI Engineer, shaping how LLMs and agentic systems are applied across the product, from prototyping intelligent workflows to building natural language interfaces. Expect end-to-end ownership, defining the playbook rather than inheriting one.

The work

• Define the AI strategy and roadmap, identifying high-leverage applications for LLMs and agentic systems across the product.
• Design and build backend services for AI workflows, from initial prototypes to production-ready systems.
• Develop and optimise prompt engineering strategies for complex tax compliance scenarios.
• Build robust evaluation pipelines to ensure model accuracy, safety, and reliability in a regulated domain.
• Architect and implement internal and customer-facing agent platforms.

What You Bring

• 5+ years building customer-facing products with end-to-end ownership in a high-quality software environment.
• Deep, hands-on experience with LLMs, prompt engineering, and agent frameworks, including production experience with RAG architectures.
• Proven ability to design and implement evaluation pipelines for AI models.