TL;DR
AI Product Manager (LegalTech): Leading the development of core models powering intelligence platform with an accent on domain-specific entity extraction and model lifecycle management. Focus on building robust evaluation frameworks, optimizing LLM performance, and driving business impact through advanced AI/ML strategies.
Location: Must be based in San Francisco and able to work from the office at least 3 days a week
Salary: $177,650 – $236,000
Company
A fast-growing vertical SaaS company backed by top VCs, dedicated to closing the justice gap by empowering law firms with AI-driven intelligence.
What you will do
- Define the strategic roadmap and tactical execution for core legal and medical entity extraction models.
- Build robust evaluation frameworks and gold-standard datasets to ensure reasoning accuracy.
- Partner with ML engineers, data scientists, and subject matter experts across the full model lifecycle.
- Balance technical trade-offs between model size and reasoning quality for commercial viability.
- Research and experiment with emerging modeling techniques to maintain a competitive edge.
Requirements
- 5+ years of professional experience in AI modeling or data science product management.
- Deep understanding of LLMs, evals, RAG, agentic orchestration, and classical machine learning.
- Proven ability to manage data quality as a first-class product and build training pipelines.
- Strong skill in translating complex model metrics into clear business impact.
- Must be based in the San Francisco area to support a hybrid work schedule.
Culture & Benefits
- Comprehensive medical, dental, and vision insurance for employees and families.
- 401(k) plan for US employees.
- Flexible paid time off, sick leave, and paid parental leave.
- Home office stipend provided.
- Regular in-person local team meet-up program.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →