Prompt Engineer

Job Title: Prompt Engineer / LLM Context Engineer

Location: London - 3 days onsite

Employment Type: Full-Time (Permanent)

We are seeking a hands-on Prompt Engineer / LLM Context Engineer to design, build, and iteratively refine prompts, instruction frameworks, and context architectures that govern how large language models behave within enterprise AI applications.

This is a technical engineering role — not creative writing — focused on delivering consistent, accurate, and auditable outputs from LLM systems operating in a regulated life sciences environment.

The successful candidate will work at the intersection of AI engineering, retrieval systems, and domain collaboration, ensuring model outputs are reliable, measurable, and production-ready.

Prompt & Instruction Engineering

• Design system prompts, structured instructions, and few-shot examples for production LLM use cases
• Iteratively refine prompts to resolve failures, edge cases, and domain-specific gaps
• Optimize prompt structure for consistency, determinism, and accuracy

Context Engineering & RAG

• Design and manage context pipelines for LLM applications
• Select, rank, and format retrieved content from knowledge bases and document stores
• Work closely with Retrieval-Augmented Generation (RAG) architectures to ground outputs in verified data
• Minimize hallucinations through strong context design

Evaluation Frameworks (Evals)

• Build and maintain systematic evaluation frameworks to measure prompt performance
• Define metrics for accuracy, consistency, and relevance
• Run regression testing and continuous prompt improvement cycles
• Use data-driven insights to guide prompt iteration (40–50% of role focus)

Cross-Functional Collaboration

• Partner with ML engineers, data scientists, and scientific SMEs
• Translate complex life sciences requirements into precise model instructions
• Support integration of LLM capabilities into production workflows

Prompt Governance & Compliance

• Maintain version control and documentation of prompt assets
• Ensure traceability suitable for regulated pharma environments
• Support audit readiness and reproducibility standards

• Proven hands-on experience engineering prompts for production LLM applications
• Strong understanding of:
• LLM behaviour and instruction following
• Context windows and tokenisation
• Hallucination mitigation techniques
• Practical experience with RAG (Retrieval-Augmented Generation) architectures
• Solid Python skills for scripting, testing, and automation
• Experience building or working with evaluation frameworks (“evals”)
• Systematic, metric-driven approach to prompt iteration
• Strong stakeholder communication skills

• Experience in life sciences, pharma, or regulated environments
• Familiarity with vector databases and retrieval pipelines
• Experience working with ML/AI engineering teams in production settings
• Understanding of governance requirements in regulated industries