Artem and Maksym from NewsCatcher here.

Some of you know us as we started six years ago as two freshly graduated economics students who decided to build the best news API product.

We started NewsCatcher thinking the market for news APIs was so big that we could build a self-serve platform and get millions of $29 users.

Obviously, it was a wrong assumption. We pivoted to serve enterprises and had success with it.

But we are hackers at heart, and we want to serve hackers.

We haven't used our Launch HN yet, so consider this our smoke test.

We're looking for feedback and power users rather than revenue. So, happy to provide enough credits for any HN user who finds CatchAll useful.

CatchAll is built for one thing: retrieving every matching event from the web. The use cases that fit it are ones where missing events have real consequences — funding and M&A monitoring, regulatory and compliance feeds (FDA approvals, SEC filings, policy changes), cybersecurity incident tracking, supply chain signals.

If your pipeline consumes structured records and the answer to your query is "find all of them," that's where it works. It's not the right tool for small, bounded queries that return 5 high-precision results.

The 15-minute job time is a direct consequence of the pipeline depth: analyze, fetch, cluster, validate, extract, deduplicate. You're not getting a ranked list of links; you're getting a verified record set.

Our latest benchmark run: https://newscatcherapi.com/blog-posts/web-search-api-benchma...