Lead Prompt Engineer for Scalable AI Workflows

Moonfire; Reading; Hybrid; GBP 55,000 - 60,000; Full time

Benefits offered by this job

Salary £55k–£60k

Equity partnership

Hybrid UK

Cutting-edge AI tech

Job summary

Moonfire seeks a Senior Prompt Engineer to lead design, optimization, and scaling of prompting strategies for large language models. You’ll partner with engineering to embed prompts in apps, agents, and RAG pipelines, driving performance and safety.

The role demands 5+ years in software engineering with 2+ years in production prompt engineering, expert Python/Java, and hands-on use of LangChain, LangSmith, and OpenAI Evals. Hybrid UK-based position with equity.

Qualifications

• 5+ years in software engineering with 2+ years in production-level prompt engineering.
• Strong expertise in LLMs: GPT-4, Claude, Llama, Mistral, Nova (or similar).
• Advanced programming skills in Python and/or Java.
• Experience with AI orchestration frameworks such as LangChain and LlamaIndex.
• Hands-on experience with prompt evaluation and monitoring tools, such as LangSmith, Fiddler, InspectAI, and OpenAI Evals.

Responsibilities

• Design and implement advanced, dynamic prompts tailored to business-specific use cases.
• Optimize LLM performance by reducing token consumption, latency, and hallucinations.
• Integrate prompts into agent-based systems, APIs, and RAG architectures.
• Develop prompt frameworks and reusable templates for scalable deployment.
• Establish evaluation, benchmarking, and regression testing mechanisms.
• Implement guardrails to mitigate prompt injection and unsafe outputs.
• Collaborate with product and engineering teams to translate business requirements into technical prompt logic.
• Maintain a version-controlled prompt repository with continuous improvements.

Skills

Python

Java

LLMs

Prompt engineering

Tools

LangChain

LlamaIndex

LangSmith

Fiddler

InspectAI

OpenAI Evals

Job description

Moonfire seeks a Senior Prompt Engineer to lead design, optimization, and scaling of prompting strategies for large language models. You’ll partner with engineering to embed prompts in apps, agents, and RAG pipelines, driving performance and safety.

The role demands 5+ years in software engineering with 2+ years in production prompt engineering, expert Python/Java, and hands-on use of LangChain, LangSmith, and OpenAI Evals. Hybrid UK-based position with equity.