Job Title: Prompt Engineer (Contract)
Rate: £550 - £600 per day
Duration: 6 months
Location: London, Tunbridge Wells, Ipswich or Bolton
Working Model: Hybrid (3 days onsite)
Industry: Banking, Financial Services & Insurance (BFSI)
The Role
We are looking for a Prompt Engineer to design, implement, and optimise conversational and generative AI solutions powered by Large Language Models (LLMs) on Microsoft Azure.
You will focus on prompt strategy, prompt chaining, Retrieval-Augmented Generation (RAG), and secure API integration, ensuring AI outputs are accurate, safe, explainable, and grounded in trusted enterprise data.
Key Responsibilities
- Design, test, and refine system, developer, and user prompts
- Implement prompt chaining and multi-step orchestration workflows
- Build and integrate LLM-powered features using Azure OpenAI and Azure services
- Develop and optimise RAG pipelines (chunking, embeddings, indexing, ranking)
- Integrate LLMs via REST APIs and SDKs into enterprise applications
- Define evaluation frameworks for accuracy, safety, latency, and cost
- Monitor production behaviour and continuously improve AI outputs
- Apply Responsible AI principles, data privacy, and security controls
Essential Skills & Experience
- Strong hands-on experience in prompt engineering and prompt chaining
- Solid understanding of LLM behaviour (tokens, temperature, context windows, tool calling)
- Experience deploying LLM solutions on Azure
- Proficiency with Python or C# (Node.js acceptable)
- Experience with RAG, embeddings, semantic search, and vector databases
- REST API integration and JSON
- Knowledge of prompt injection risks and AI security considerations
- Git and CI/CD experience (Azure DevOps or GitHub)
Desirable
- LLM evaluation frameworks and prompt quality metrics
- MLOps and observability for LLM applications
- Document processing and ETL for retrieval pipelines
