Back to all vacancies

AI Engineer

London, SE1 8NWPosted: 15th September 2026

Job Purpose

The AI Engineer is responsible for building and improving the components that make up the intelligence layer of Croner’s generative AI solutions. This includes implementing Retrieval-Augmented Generation (RAG) pipelines, agentic workflows, prompt strategies, and evaluation harnesses that deliver accurate, trustworthy, and domain-specific AI behaviour.

As part of Croner’s AI team, the role turns agreed designs and technical direction into working, production-ready implementations. The AI Engineer works to the patterns and standards set by the Senior AI Engineer and the Director of Search & Generative AI, and collaborates closely with the AI Platform Engineer—who owns infrastructure, runtime, and operational concerns—so that the capabilities they build integrate cleanly with the wider platform.

This is a hands-on delivery role with clear scope for growth. The AI Engineer is expected to take ownership of well-defined features from implementation through to release, to raise questions and blockers early, and to build depth across retrieval, model behaviour, and evaluation as Croner’s AI capabilities mature.

 

Objectives

  • Deliver well-scoped generative AI features to agreed quality standards, focused on correctness, faithfulness, and robustness.

  • Implement and improve RAG and agentic components that make effective use of Croner’s proprietary content.

  • Build and maintain evaluation and experiment tooling that makes AI performance measurable.

  • Apply prompting techniques and, with guidance, model adaptation techniques to improve quality, latency, and cost.

  • Produce clear, tested, well-documented code that colleagues can pick up and extend.

  • Grow technical depth in generative AI engineering, actively seeking review and feedback from senior colleagues.

Key Responsibilities

1. RAG & Generative AI Engineering

  • Implement generative AI workflows using LlamaIndex, AWS Strands Agents SDK, and similar frameworks.

  • Build and maintain components of RAG pipelines, including retrieval logic, prompt templates, grounding strategies, and response structuring.

  • Implement agentic or workflow-based patterns to agreed designs, with attention to determinism, explainability, and maintainability.

2. Retrieval & Knowledge Engineering

  • Implement and tune retrieval strategies such as chunking approaches, reranking logic, and retrieval fusion, guided by evaluation results.

  • Support work on advanced retrieval techniques, including Graph-RAG and structured knowledge representations.

  • Work with the AI Platform Engineer so that retrieval logic fits the vector search, indexing, and ingestion pipelines in use.

3. Model Adaptation & Optimisation

  • Develop and iterate on prompts and model configurations to improve answer quality.

  • Support fine-tuning, instruction tuning, and knowledge distillation work, including dataset preparation, training runs, and comparison of results.

  • Help assess trade-offs between prompt-based approaches and model adaptation in terms of quality, latency, and cost.

4. Evaluation, Experimentation & Quality

  • Build and run evaluation workflows for generative AI features, using tooling such as MLflow.

  • Assemble and curate test sets, and measure retrieval quality, answer correctness, faithfulness, and robustness.

  • Keep experiments reproducible, documented, and ready to fold into wider evaluation and monitoring frameworks.

5. Engineering Practice & Quality

  • Write clean, tested Python that meets the team’s standards for readability, structure, and documentation.

  • Take part in code reviews and design walkthroughs, both giving and acting on feedback.

  • Follow and help improve the team’s practices for version control, CI/CD, and experiment tracking.

6. Collaboration

  • Work day to day with the Senior AI Engineer, taking technical guidance and escalating blockers early.

  • Partner with the AI Platform Engineer to move AI logic from experimentation into production.

  • Work with product and delivery colleagues to clarify requirements and support delivery.

Skills & Requirements

Technical Skills

  • Solid proficiency in Python, with a working understanding of what production-grade code requires.

  • Hands-on experience with at least one generative AI framework, such as LlamaIndex, LangChain / LangGraph, or the AWS Strands Agents SDK.

  • Practical experience implementing or contributing to Retrieval-Augmented Generation (RAG) systems; exposure to Graph-RAG is a plus.

  • Working understanding of evaluation and experimentation methods for generative AI systems; experience with MLflow or similar is advantageous.

  • Exposure to fine-tuning, instruction tuning, or knowledge distillation is desirable rather than essential.

  • Familiarity with AWS, Bedrock, AgentCore, or Databricks environments is advantageous.

  • Competence with code versioning and a working understanding of CI/CD pipelines.

Experience

  • Demonstrable experience delivering software, data science, or machine learning solutions in a professional environment.

  • Demonstrable experience contributing to systems that have reached production, including an appreciation of testing, monitoring, and operational concerns.

  • Demonstrable experience taking ownership of well-defined features or components, from implementation through to release.

  • Demonstrable experience working with real datasets and data pipelines, including basic considerations of performance and reliability.

Behavioural Skills

  • Strong interest in generative AI and a clear appetite to keep learning in a fast-moving field.

  • Strong problem-solving skills and care for implementation quality.

  • Comfortable asking questions, sharing work early, and acting on feedback.

  • Able to balance experimentation with the need to deliver.

  • Clear and effective communicator within technical and cross-functional teams.

Interested in this role?

Grouprecruitmentuk@peninsulagrouplimited.com

Find us on: