Evaluation Engineer

Freshfields Bruckhaus Deringer; England; On-site; GBP 65,000 - 95,000; Full time

Job summary

Freshfields Bruckhaus Deringer seeks a QA Engineer to build and run tests for AI applications, evaluating accuracy, citations, and retrieval relevance. You will prepare evaluation datasets with legal specialists, ensure confidentiality, and follow stringent data-use rules.

You will work with senior engineers to implement automated test cases, maintain regression suites, and analyse failures across model and application versions, contributing to robust AI systems for legal tasks.

Qualifications

• Five years’ software development experience with Python, Java or C#.
• Foundational understanding of large language models and their failure modes.
• Ability to use coding agents and best practices for prompting and security.
• Experience with APIs, Git, automated testing and CI/CD tools.
• Curiosity to learn in a rapidly evolving AI landscape.

Responsibilities

• Develop automated evaluation cases and supporting scripts.
• Prepare test datasets with legal specialists and ensure data confidentiality.
• Run evaluations of answer accuracy, citations, and retrieval relevance.
• Execute functional tests for AI workflows with QA collaboration.
• Review outputs to identify failures and reproduce issues.
• Compare results across model, prompt and application versions.
• Assist calibration of automated graders with human reviews.
• Maintain regression suites and CI/CD integrations.

Skills

Python

Java

C#

LLMs understanding

Testing frameworks

CI/CD concepts

Prompting best practices

Documentation & debugging

Tools

Git

Jira

Jenkins

Job description

Role purpose

Build and run tests that assess the quality and reliability of the firm’s AI applications. Help maintain evaluation datasets, investigate failures and provide clear evidence of how AI systems perform on legal and business tasks.

Key responsibilities

• Implement automated evaluation cases and supporting scripts using established frameworks and guidance from senior engineers.
• Prepare and maintain test datasets, source documents and reference examples with legal specialists. Follow requirements for confidentiality, permitted data use and version control.
• Run evaluations of answer accuracy, completeness, citations, retrieval relevance, instruction following and appropriate escalation. Apply agreed scoring criteria and flag cases requiring expert judgement.
• Execute functional and integration tests for AI workflows, including document ingestion, tool calls, structured outputs and error handling, in coordination with application QA.
• Review outputs and execution traces to identify failures. Reproduce issues, document supporting evidence and help distinguish data, retrieval, model and application defects.
• Compare results across model, prompt and application versions. Report changes accurately, including uncertainty and limitations, rather than relying solely on headline scores.
• Support the calibration of automated graders by comparing their scores with human reviews and investigating inconsistencies.
• Maintain regression suites, contribute to CI/CD integration and turn approved production failure examples into new test cases.
• undefined

Required skills and experience

• Minimum five years’ experience developing software demonstrated through employment, placements or substantial projects using programming languages such as Python, Java or C#.
• Foundational conceptual understanding of large language models, including their capabilities, limitations and common failure modes, and an ability to apply that understanding when developing evaluation criteria, test cases and acceptance thresholds with guidance from senior engineers.
• Ability to use coding agents effectively to support development, testing and analysis, with awareness of best practices for prompting, context management, validation, security, responsible use of generated code and when to seek human review.
• Working knowledge of APIs, Git, automated testing and software development lifecycle practices and CI/CD tools such as Jira and Jenkins.
• Demonstrated curiosity and ability to learn in a rapidly evolving AI landscape, assess new models, tools and evaluation techniques, and apply relevant learning pragmatically with appropriate guidance.
• Foundational understanding of statistics and measurement, including sampling, precision and recall, together with strong analytical skills, attention to detail and the ability to document issues clearly and seek technical or legal guidance when needed.
• Practical experience evaluating AI applications is a plus but is not required, given the newness of the field; relevant academic, internship or project experience will also be considered.