Artificial Intelligence Engineer

SearchWorks; Greater London; On-site; GBP 90,000 - 125,000; Full time

Job summary

SearchWorks in London is seeking a Senior AI Engineer to design, build and scale production-ready LLM and agentic systems across the business. You will collaborate with underwriters and analytics teams to translate complex workflows into practical, secure, scalable AI solutions hosted on AWS, using React/TypeScript on the frontend.

The role focuses on turning experiments into shipped products with measurable business impact, from data retrieval to multi-step orchestration.

Qualifications

• Commercial software engineering experience across backend and full-stack.
• Hands-on React and TypeScript development.
• Strong knowledge of AWS cloud-based app development.
• Practical experience building or integrating LLM/generative AI.
• Track record delivering secure, production-ready software.

Responsibilities

• Designing, building and scaling production-ready LLM applications.
• Developing full-stack products using Python, React and TypeScript.
• Creating AI-assisted workflows for document processing, data retrieval and automation.
• Building secure backend services and AWS integrations.
• Designing retrieval pipelines with embeddings and vector databases.
• Developing agentic workflows with tool calling and orchestration.
• Collaborating with stakeholders to translate complex requirements into technical solutions.
• Improving reliability with grounding, evaluation, observability and structured outputs.

Skills

Software engineering

React

TypeScript

AWS

LLM / Generative AI

Stakeholder communication

Tools

AWS Lambda

S3

PostgreSQL

pgvector

OpenAI / Bedrock

Vector search

SQS/SNS

Job description

The opportunity - Senior AI Engineer - LLM & Agentic Systems

A global insurance and financial services organisation is expanding a specialist LLM engineering team responsible for delivering production-ready AI applications across the business.

The team operates like a startup within an established enterprise: fast-moving, experimental and commercially focused. You will work directly with underwriters, analytics teams, operations leaders and other subject-matter experts to understand complex workflows and turn ambiguous opportunities into secure, scalable systems with measurable business impact.

This is an opportunity to build practical LLM applications—not isolated prototypes. The team has already developed AI-assisted tools that automate workflows and make internal business data easier to find and use. You will help scale these products and develop new capabilities across the organisation.

What you’ll be doing

• Designing, building and scaling production-ready LLM applications
• Developing full-stack products using Python, React and TypeScript
• Creating AI-assisted workflows for document processing, data retrieval and business automation
• Building secure backend services and integrations on AWS
• Designing retrieval pipelines using embeddings, semantic search and vector databases
• Developing agentic workflows involving tool calling, function execution and multi-step orchestration
• Working directly with business stakeholders to turn complex requirements into practical technical solutions
• Improving the reliability of LLM applications through grounding, evaluation, observability, fallback logic and structured outputs
• Supporting products from early experimentation through to production deployment

What we’re looking for

• Strong commercial software engineering experience across backend and full-stack development
• Experience working with React and TypeScript
• Strong knowledge of AWS and cloud-based application development
• Practical experience building or integrating LLM or generative AI applications
• A track record of delivering secure, maintainable and production-ready software
• Strong communication skills and the ability to work directly with technical and non-technical stakeholders

Particularly relevant experience

• AWS Lambda, S3, SQS, SNS, Textract or Secrets Manager
• PostgreSQL, pgvector, embeddings, vector search or semantic retrieval
• OpenAI, AWS Bedrock, prompt engineering or LLM orchestration
• Agentic AI, tool calling, function execution or human-in-the-loop workflows
• RAG and AI-assisted document-processing systems
• Structured outputs, grounding, evaluation and prompt-version management
• Snowflake, Microsoft Graph or enterprise identity platforms
• SignalR or other real-time application patterns
• Insurance, reinsurance, fintech or another regulated environment