Legal Knowledge Engineer - Transactional

About the role

The Legal Knowledge Engineer - Transactional is an integral part of Latham’s Legal Innovation team. This role will develop and curate the knowledge foundations powering AI-enabled legal workflows, and create and maintain evaluation datasets, test cases, and quality benchmarks for AI-powered tools.

This role is based in either Latham’s New York, Boston, or Silicon Valley offices. The position may offer a flexible working schedule allowing for a hybrid and in-office presence.

Responsibilities

• Develop and implement quality assurance protocols and evaluation frameworks for AI outputs, establishing standards for accuracy, completeness, and alignment with attorney expectations.
• Translate transactional practice requirements into structured knowledge repositories, prompt templates, and context specifications that optimize AI system inputs for high-quality outputs.
• Analyze AI output quality across workflows by identifying patterns, gaps, and opportunities to improve context, inputs, and evaluation criteria.
• Monitor and report on quality metrics, accuracy measures, and evaluation results for AI-powered workflows, using data-driven insights to guide knowledge asset refinement and context optimization.
• Conduct research to identify best practices for knowledge curation, context management, and AI evaluation methodologies, while staying current with advances in retrieval-augmented generation and legal AI applications.
• Protect and maintain any highly sensitive, confidential, privileged, financial, and/or proprietary information that Latham & Watkins retains.

Qualifications

• Possess strong expertise in knowledge engineering and AI systems, with the ability to curate and structure information to optimize AI inputs and outputs for accuracy, consistency, and alignment with legal practice standards.
• Demonstrate a solid understanding of transactional legal practice, including mergers and acquisitions (M&A), finance, and capital markets workflows, with the ability to translate practice expertise into structured knowledge assets and evaluation criteria.
• Exhibit strong knowledge curation capabilities, including developing reference libraries, context repositories, evaluation datasets, and quality assurance frameworks for AI-powered transactional tools.