# Harvey — How We Built Agentic Retrieval for Patent Research

- Company: Harvey (harvey.ai)
- Announced: 2026-11-19T16:00:00+00:00
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://www.harvey.ai/blog/how-we-built-agentic-retrieval-for-patent-research
- Record: https://forck.live/items/38699-how-we-built-agentic-retrieval-for-patent-research
- Subject: Harvey

Harvey's patent research agent navigates fragmented patent sources, deciding what to seek next and conducting focused research within a bounded budget. The company says a single patent index falls short because patent information is scattered across published patents, patent-office systems, indexes and web sources, each with its own boundaries. Traditional retrieval-augmented generation searches a fixed index and works when the information sits in one collection and the research path is predictable, but in patent research a result may reveal another record to inspect or another fact to verify. Each task starts with a research objective that preserves the user's full question, including patent numbers, claim numbers, timing and legal context, so the agent plans from the desired outcome rather than a reduced bag of keywords. For recurring patterns such as current claim status, art applied by an examiner, PTAB outcomes and prior-art research, Harvey gives the agent concise paths, which it describes as information requirements rather than hard-coded source assignments. The agent receives a routing table listing what data each source has, which information needs it serves best, where its boundaries lie and which alternative can be a fallback; it prioritizes the USPTO for U.S. patent text and names Google Patents as an alternative when the USPTO isn't available. The agent stops when it can return a cited answer, identify what remains unresolved, or when it reaches the run's research budget.

## Evidence

Verbatim from https://www.harvey.ai/blog/how-we-built-agentic-retrieval-for-patent-research:

> In Harvey, an agent navigates fragmented patent sources, decides what information to seek next, and conducts focused research within a bounded budget.
> 
> Patent research often requires more than finding a document. A question about a claim may depend on its history: how it changed during examination, how the applicant addressed prior art, and whether later proceedings affected its status.
> 
> This is the central challenge of patent research: the answer often depends on following a record over time and deciding what to inspect next . Patent attorneys do this naturally. They start with one document, notice a later event or missing fact, open another record, and continue until the question is supported, or until they can clearly identify the remaining gap.
> 
> Consider a question that sounds simple: What does claim 8 cover? A conventional retrieval system can find the patent and return claim 8 as issued. But depending on what the attorney needs to know, the question may involve:
> 
> What did the application originally claim?
> 
> What language appeared in the issued patent?
> 
> How did the applicant distinguish the claim from prior art?
> 
> Is the claim still in force today?

---

Record: https://forck.live/items/38699-how-we-built-agentic-retrieval-for-patent-research
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
