Description
As a Sr. Legal AI Prompter / Evaluator, you will play a crucial role in improving CoCounsel Legal by writing and refining prompts, skills, and agent configurations, experimenting with large language models (LLMs), and evaluating the results.
You will partner closely with specialized legal subject matter experts (SMEs), software engineers, and data scientists to ensure that products deliver accurate, legally sound outputs that legal professionals can trust.
Key responsibilities include:
- Creating, testing, and refining prompts, skills, and agent configurations for CoCounsel Legal
- Experimenting with Thomson and other LLMs on legal tasks, comparing results to inform which model best fits each use case
- Building gold-standard datasets with SMEs, coordinating evaluations, and designing evaluation schemes
- Writing and running scripts, experiments, and automated tests to ensure quality and consistency
- Monitoring production performance, analyzing failure modes, and providing strategic feedback
- Collaborating with software engineers, data scientists, and cross-functional teams
- Engaging with legal professionals, industry leaders, and internal stakeholders to gather insights and validate product concepts
- Mentoring colleagues and drafting guidance materials on getting the most out of LLMs
The ideal candidate has a Juris Doctor degree or equivalent law degree and bar admission in a US or Canadian jurisdiction, 5-7 years of combined experience, and proven expertise applying generative AI.
In terms of benefits, Thomson Reuters offers a hybrid work model, flexibility and work-life balance, career development and growth opportunities, industry-competitive benefits, and a culture that promotes inclusion and belonging.
The base compensation range for this role varies by location: $137,900 USD - $256,100 USD for eligible office locations in the US, $119,700 USD - $222,300 USD for other US locations, and $126,000 CAD - $176,000 CAD for Ontario, Canada.
For more fresh new jobs, check out yubhub.co.