Prompting Engineer
Are you passionate about shaping the future of Artificial Intelligence? We are seeking a skilled AI Prompt Engineer to join our dynamic Commercial Digital team. In this role, you will take ownership of the behavior of Large Language Model (LLM)-driven agents in production.
Working within an established framework, you will develop and tune prompts, design evaluations for individual AI agents, and implement self-optimizing feedback loops. Your mission is to ensure our customer-facing AI outputs remain highly accurate, on-brand, legally compliant, and cost-efficient.
Key Responsibilities
- Develop & Tune Prompts: Craft and maintain high-performing prompt architectures—including system prompts, tool instructions, guardrails, and output schemas—for production AI agents.
- Evaluate & Test: Build and execute evaluation suites and regression tests for individual agent behavior. Define quality metrics and acceptance thresholds in collaboration with business owners.
- Optimize Performance: Design and operate A/B testing for prompt and model variants, utilizing feedback loops that learn from human review outcomes to focus on single-agent optimization.
- Drive Cost Efficiency: Support LLM optimization efforts, focusing on model selection, context-window management, structured outputs, and token efficiency.
- Guard Quality & Tone: Maintain tone-of-voice and multi-language consistency for customer-facing communications, utilizing professional British English B2B standards as the baseline.
- Risk Mitigation: Stress-test guardrails against adversarial inputs, edge cases, and hallucination risks within defined safety standards before prompt release.
- Document & Collaborate: Document prompt patterns, test coverage, and evaluation results, enabling business users and reviewers to provide structured feedback.
Reporting Line & Autonomy
- Reporting: You will report directly to the Head of Commercial Digital, receiving day-to-day technical guidance and mentorship from our Level A Senior Prompt Engineers.
- Autonomy: You will work within defined organizational standards. Production prompt changes will be peer-reviewed before deployment to ensure quality assurance.
Requirements & Qualifications
- Experience: 1+ year of practical, hands-on experience working with LLM applications (including prompting, Retrieval-Augmented Generation (RAG), and structured outputs).
- Track Record: Proven exposure to systematic testing, data validation, or Quality Assurance (QA) of AI/ML systems.
- Technical Skills: Strong prompt-writing craft, disciplined experimentation habits, and basic Python or data-handling skills to run evaluation pipelines.
- Language Skills: Exceptional written English with a strong sensitivity to tone, style, and nuance in professional B2B communication.
- Professional Attributes: Clear documentation skills, attention to detail, and a structured approach to processing user feedback.
- Preferred Skill: Previous experience with the Microsoft AI stack (Azure OpenAI, Azure AI Foundry, or Copilot Studio) is a significant advantage.
Our Values & Expectations
We expect our team members to drive growth, embrace innovation, and inspire others. You will succeed in this engineering role if you align with our core behaviors:
- Inspire Your People: Simplify complexity, act with the long-term vision in mind, and learn from setbacks with high integrity.
- Strengthen Your Team: Set the bar high for AI quality, collaborate across departments, and foster a growth mindset.
- Grow Your Business: Put customers at the heart of your prompt designs and proactively look for ways to improve automated processes.
Our Core Values:
- Own It: We believe in teamwork, keeping promises, and building our brands together.
- Aim High: We are bold, positive, and strive to perform at our best.
- Be True: We act with absolute integrity to win for our stakeholders, colleagues, and the environment.
About us - Malvern Panalytical ,