Aadil Manazir
14 posts



Pushing the Harvey for Word Add-In further exposed the limits of .docx. So Harvey’s engineers designed an agent that builds documents in memory, offloading state to the backend so the model can focus on legal text. 40% higher edit acceptance. 70% more usage.

In legal work, outcomes depend on the quality and coverage of your knowledge sources. In this blog, @samarth__go and Christopher Bello explain how we built the Data Factory to discover authoritative legal sources, validate them for compliance, and test real legal reasoning at scale. The result: Harvey scaled from six jurisdictions to 60+, and from 20 legal data sources to 400+. Full breakdown: harvey.ai/blog/using-age…





By popular request: you can now branch conversations in ChatGPT, letting you more easily explore different directions without losing your original thread. Available now to logged-in users on web.








