SWE, Legal Prompter

Software Engineer, Legal Prompter
Are you looking for an exciting new opportunity in 2024? Thomson Reuters is looking for a Legal Prompter to join our Machine Learning Engineer team, working on CoCounsel.
About the Role
We are seeking a creative backend engineer to join the team behind CoCounsel. You will help build applications that interact with state-of-the-art Large Language Models (LLMs) that deliver cutting-edge and time-saving solutions to attorneys all around the country. CoCounsel is already capable of writing legal research memos, summarizing and reviewing documents, and red-lining contracts, just to name a few capabilities–all in minutes, not days or weeks. And were just getting started.
As a Software Engineer, Legal Prompter, you will develop Python code as well as English prose prompts to LLMs to help expand CoCounsels early success. This mix of writing instructions for an interpreter and an LLM is well-suited for someone who can produce reliable backend code, and can also write creative, lucid, and effective prose instructions.
Our ideal candidate is a software engineer who can write high-quality Python, work with APIs (consuming and serving), and also has domain knowledge of the law (JD, legal industry experience, etc.). A legal background is absolutely necessary.
As the Software Engineer, Legal Prompter, you will:
Orchestrate LLM-driven experiences that delight our customers
Develop Python code that interacts with state-of-the-art LLMs
Write, test, and iterate prose prompts for the LLM that leverage your legal domain knowledge, ensuring delivery of accurate, legally sound, and contextually appropriate responses
Develop Python code that interacts with data sources & services (APIs, SQL databases, etc.)
Liaise with legal professionals and other teams at Casetext to gather and understand requirements
Integrate the language models into multi-step systems, assessing their performance and making necessary modifications
About You
Legal domain expertise. This can mean experience in legal practice, a JD-holder in a “non-traditional” legal career path, a paralegal, or legal industry experience
One to two years of production level experience in software engineering on an engineering/product team, preferably at a startup
Proficiency in Python – if Python skills are low, then strong experience in other programming languages is necessary
Proficiency with APIs and SQL is a major plus
Excellent analytical and problem-solving skills
Ability to work both independently and in a collaborative environment
Creative and clear writing skills
Experience using large language models through code is preferred (such as model services from OpenAI, Anthropic, or Cohere; and/or open source models such as Llama and Mistral)
#LI-ES1
Whats in it For You?
You will join our inclusive culture of world-class talent, where we are committed to your personal and professional growth through:
Wellbeing: Comprehensive benefit plans; flexible and supportive benefits for work-life balance: flexible vacation, two company-wide Mental Health Days Off; work from another location for up to a total of 8 weeks in a year, 4 of those weeks can be out of the country and the remaining in the country, Headspace app subscription; retirement, savings, tuition reimbursement, and employee incentive programs; resources for mental, physical, and financial wellbeing.
Culture: Globally recognized and award-winning reputation for equality, diversity and inclusion, flexibility, work-life balance, and more.
Learning & Development: LinkedIn Learning access; internal Talent Marketplace with opportunities to work on projects cross-company; Ten Thousand Coffees Thomson Reuters cafe networking.
Social Impact: Ten employee-driven Business Resource Groups; two paid volunteer days annually;