Hey everyone! 👋

We're Abby and Cody, co-founders of Riveter (F24). We both did 5+ years at Gusto, followed by Middesk and Retool, and watched the same painful problem play out across eng, data, ops, and growth teams: the information you need exists online, but getting it into a structured, usable dataset is brutally hard.

TL;DR: Write a prompt. Riveter builds the list and enriches it -- scraped live from the web. No stale databases, no stitching tools together.

Teams at OpenAI, Snap, and Fermat are already using Riveter to build datasets they couldn't justify building manually.

Ask: We'd love to show you what Riveter can do for your use case. Book 20 minutes with us and we'll give you 1,000 free credits to explore the platform.

https://youtu.be/KH0IELrOtLs

❌ The Problem

Almost any list you could ever need can be derived from the web. Every law firm in California. Every vet practice by state. Every competitor's pricing page. The information is out there.

The problem is getting it out.

Current options all hit the same wall:

• Data vendors sell you records from a static index -- outdated the moment you buy them, and missing anything outside their database
• Clay and Parallel pull from the same stale indexes. When we asked both to return all YC Winter 2026 companies, Parallel returned an incomplete dataset at a way higher cost. Clay returned "Y Combinator" as a company, along with results from other batches.
• ChatGPT and Claude can reason about the web but can't handle the scale -- you can't ask an LLM to return thousands of enriched records reliably

The result: teams buy incomplete data, spend weeks building it manually, or go without.

✅ Our Solution

Riveter agents navigate the web the way a researcher would -- running searches, visiting pages, reading results -- and return a fully structured, enriched dataset.

Write one prompt. Get thousands of rows.

Describe what you want: "Return every dental practice in San Francisco, the dentists at each practice, and their contact details"

1. Riveter agents search the web, visit pages, and compile the list -- live, not from a cached index
2. Custom enrichments get layered on: lead scores, pricing data, contact info, whatever your workflow needs
3. The result lands as a structured dataset, ready to use or pipe into your existing tools

No code required. Use it in-app, via API, or through our MCP server.

🔍 Use Cases

Sales and lead generation: Find hyper-specific lists no vendor has pre-built. Every locksmith in the US. Every independent law firm in California. Every vet practice by state. Add enrichments in Riveter to qualify, score, and route them into your outreach.

Product data: School logos, all attorneys at every AmLaw 100 firm, every menu item at every restaurant chain in a market. If it's on the web, Riveter can build it.

Market intelligence: Track competitors and their pricing on a daily or weekly cadence. Especially common for e-commerce teams who need to move fast when something shifts.

🙏 Our Ask

• Book 20 minutes with us - tell us what you're building and we'll set up your first dataset together, plus 1,000 free credits on us
• Try it yourself at riveterhq.com - free to get started in-app, via API, or MCP.
• Know a team paying too much for stale data or drowning in manual research? Send this their way.

👋 The Team

Abby Grills (CEO) - Previously at Gusto and Middesk across product and data workflows. Riveter is the tool she kept wishing existed.

Cody Watters (CTO) - Previously at Gusto and Retool. Watched too much engineering time disappear into one-off data pipelines that should have been solved at the platform level.

We built Riveter because we kept hitting the same wall: the data we needed was obviously on the web, and there was no good way to get it out at scale.