About the role
About Crosby AI Crosby is an AI-first legal platform reimagining corporate legal services from the ground up. We are a team of technologists and legal experts who build proprietary technology and human-in-the-loop workflows that meaningfully improve how lawyers and machines work together — delivering speed, consistency, and quality across high-stakes work. Our systems review complex documents faster and with exceptional accuracy, combining advanced AI with structured legal expertise. Clients receive AI-powered redlines, commentary, and negotiation guidance within hours, at a predictable, volume-based price. Backed by Sequoia, Index Ventures, and Bain Capital Ventures, we're building the end-to-end contracting platform for the next generation of fast-growing companies.
The Team The Engineering team builds the core AI systems and infrastructure that power Crosby’s platform. We work at the intersection of research and production, developing models and systems that operate reliably in complex, real-world legal workflows. We prioritize rigor, speed, and impact — advancing the frontier of applied AI while shipping systems that deliver measurable value.
The Role As an AI Researcher at Crosby, you'll play a critical role in advancing the capabilities of our AI systems. You’ll explore, develop, and apply state-of-the-art techniques in LLMs and applied NLP to improve how our platform understands, analyzes, and reasons over complex legal documents. You’ll work closely with engineers, data scientists, and legal experts to translate research ideas into production systems. This role is for someone who is deeply curious, technically rigorous, and motivated to push the boundaries of what AI can do in high-stakes domains.
Responsibilities Advance model capabilities: Research and develop novel approaches to improve performance on tasks like document understanding, clause extraction, reasoning, and negotiation. Experiment and iterate: Design and run experiments to evaluate new architectures, prompting strategies, fine-tuning approaches, and retrieval methods. Bridge research and production: Work closely with engineering teams to translate research insights into scalable, production-ready systems. Improve evaluation: Develop benchmarks, datasets, and evaluation frameworks to rigorously measure model performance. Leverage modern AI techniques: Stay at the frontier of LLMs and applied AI, incorporating new methods where they meaningfully improve outcomes. Collaborate cross-functionally: Partner with engineering, product, and legal teams to ensure research efforts align with real-world needs.