15k pages of discovery: Claude Code finds the evidence, with a trail
About 15k pages of discovery stood between a solo practice and a summary judgment motion. Claude ran OCR, built embeddings and search scripts, then found evidence for each assertion, with traces to check.
Done withClaude CodeSpotted· by HN user droidjj· Oct 22, 2026
Result Trusted because it left traces to follow
“I trusted it because there were traces I could follow.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
As a solo litigator, they had around 15k pages of discovery, almost entirely PDFs, to dig through for a summary judgment motion. Instead of asking Claude outright to find the best evidence, they first had it run a clean, high quality OCR pass, then generate embeddings and reusable python scripts for keyword and semantic search.
How it went
While writing the brief, they would routinely ask Claude Code to use both keyword and semantic search to find the best evidence for whatever assertion they were making, and trusted it because there were traces they could follow. They note that just handing a model all the documents and asking it to write a brief arguing a position produces evocative, low signal to noise writing instead.
Try it yourself with Claude Code
I have a large set of discovery documents (mostly PDFs) for [a legal motion/brief]. Run a clean, high quality OCR pass on them, then generate embeddings and build reusable keyword and semantic search over the results. When I'm drafting, search for the best supporting evidence for a specific factual assertion and show me where it came from.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
Before a conference with 40 people they had never met, LinkedIn research normally takes an evening. Claude Code researched the attendees in parallel via LinkedIn and Chrome MCP, producing a full briefing in 5 minutes.
Surveys, supplier risk, emissions reports, modern slavery checks: the data gathering never ends. With Claude Code they built tools that now do about 70% of it. Plus a game of Frogger, for breaks.
Markdown files should just open, so they asked Claude Code for a viewer. It opens instantly: faster, they say, than any other macOS app, and faster than Bitwarden's Chrome extension shows its fill menu.
Migrating a password reset flow, Claude Code and Opus produced code on par with a mid level developer. It needed close supervision throughout, and cost roughly 2 months of Pro subscription for about half a day's work.
A side project needed a launch video, but not an agency's bill. Claude Code built one from the codebase plus their feedback. It took a lot of iteration over 2-3 days, all on subscription token usage.
A URL shortener side project ran on AWS. Claude Code wrote the GitHub Actions workflows for the move and fixed each failure as it came up. The writer reports zero downtime.