TechPrufer
Get started
Harvey
Harvey logoHarvey · Feature trail

Firm Knowledge Search

6 posts · Dec 2025 – Jul 2026

What this trail covers

Harvey's platform continues to evolve its AI-powered legal research capabilities, focusing on the integration and reasoning over authoritative legal sources. The system leverages advanced AI to move beyond traditional keyword-based searches, providing synthesized, reasoned answers grounded in primary and secondary legal content. This evolution emphasizes source transparency, verification, and the ability to handle complex legal questions efficiently. The platform aims to complement, rather than replace, human expertise by enabling users to easily find and work with firm-specific knowledge, past matters, internal guidance, and expertise. This includes permission-aware search, intelligent summarization of findings, and integration with various data sources via native integrations, APIs, and MCP.

All Harvey trails
  1. AI Trademark Search: From Clearance to Brand ProtectionThis post details the application of AI to trademark search and brand protection. It describes how AI trademark search uses machine learning to identify potential conflicts by matching sound, meaning, and visual similarity across trademark registries. The post highlights the limitations of AI in making the final legal judgment (likelihood of confusion) and emphasizes the role of legal professionals. It explains how AI assists in analyzing search results, drafting clearance opinions, responding to office actions, and monitoring/enforcing marks. Specifically, it details how AI digests large search reports, organizes marks by risk, maps them against confusion factors, and drafts initial opinions. For office actions, AI reads examiner reasoning, pulls precedent, and drafts structured responses. For enforcement, AI helps sort threats and drafts opposition analyses and cease and desist letters. The post contrasts trademark search databases with legal AI platforms, positioning Harvey as a legal AI platform that picks up where search ends, working with search results to analyze, draft, and review legal correspondence. It also mentions the integration of client and brand data under enterprise-grade security and the consistency of reasoning across different legal tasks.
  2. Search That Works With Firm KnowledgeThis post details the evolution of Harvey's Firm Knowledge search capability, emphasizing its ability to securely connect and surface relevant firm-specific information. Key technical aspects include permission-aware search, intelligent summarization of findings, and integration with various data sources (DMS, CLM, email) through native integrations, APIs, and upcoming MCP support. The post highlights use cases like streamlining due diligence, accelerating onboarding, and unlocking new revenue streams by making proprietary knowledge discoverable and usable at scale. It also reiterates the importance of enterprise-grade governance, including authentication, audit logs, role-based access, and ethical walls.
  3. How Agentic Search Unlocks Legal Research Intelligence at HarveyThis post details the implementation of 'Agentic Search,' a system that moves beyond naive RAG to an iterative reasoning process for legal research. Key technical contributions include: 1. **Agentic Search Workflow:** An implementation inspired by ReAct, interleaving reasoning and tool calls for query understanding, dynamic tool selection, retrieval, synthesis, completeness checks, and citation-backed responses. 2. **Addressing Key Challenges:** Solutions for 'Source Selection in the Wild' (improving tool selection precision to 0.8-0.9) and 'Scaling Tool Calls to Query Complexity' (calibrating agent effort to scale to 3-10 retrieval operations for complex queries). 3. **Privacy-Preserving Evaluation:** A data flywheel involving expert query generation, rapid prototyping, systematic evaluation using LangSmith traces, and targeted improvements to tool design, system prompts, and tool docstrings. This process enabled significant improvements in accuracy and efficiency for complex, multi-source legal queries.