{"id":99812,"title":"CatchAll: Recall-first web search API ","tagline":"Turning the web into a structured database of real-world events","body":"Hello, hackers! We’re Artem and Maksym, founders of NewsCatcher.\\\n\\\nWe’re building a structured, event-centric index of what happens in the real world. Today, that starts with CatchAll — a recall-first search API that finds all relevant matches to a query, extracts structured facts, and delivers clean, ready-to-use datasets.\\\n\\\n\u003chttps://www.youtube.com/watch?v=GPfVdNRoDsA\u003e\\\n\\\n**The problem**\\\nTraditional web search — and most search APIs — were designed for humans. They optimize for **speed and ranking**, which works well when one page contains the answer:\n\n* “Who is the CEO of a company?”\n* “What’s the latest acquisition in X?”\n\nBut many high-stakes questions don’t have a single answer:\n\n* _All_ regulatory actions affecting an industry\n* _Every_ cybersecurity incident in a given week\n* _All_ funding rounds, policy changes, or facility expansions\n\nIn these cases, **recall is everything**.\\\nIf 200 valid events exist and your system surfaces 5, your recall is 2.5%. No amount of prompting or post-processing fixes that.\n\nLLMs and “deep research agents” haven’t solved this. They still open and read pages one by one, hit context limits, and end up sampling rather than exhausting a topic.\n\n\\\n**The solution**\\\nCatchAll is a recall-first web search API built for long-list answers — answers distributed across the open web.\n\nCatchAll:\n\n* Pulls a large candidate set from NewsCatcher’s proprietary web index (often tens of thousands of pages)\n* Validates which pages actually match the criteria\n* Extracts structured fields (companies, dates, locations, actions, amounts — defined by the user)\n* Normalizes and deduplicates everything into clean records\n\nThe output isn’t links.\\\nIt’s a dataset that didn’t exist before — and can be kept up to date by turning the query into a **live monitor**.\n\nFor teams, the end result is fewer searches and more reliable alerts.\\\n\\\n**Backstory**\\\nMaksym and I have known each other for almost 20 years. NewsCatcher is our first startup — we've been building it since 2020. It started as a bootstrapped self-serve news API for startups, and then grew into custom enterprise solutions (powering the US Department of State, Transparency International, Samsung, 50+ enterprise clients). CatchAll is us coming back to self-serve: same infrastructure, now optimized for the AI agent ecosystem.\\\n\\\nSign up [here](https://platform.newscatcherapi.com/catchall?utm_source=launch_yc\u0026utm_medium=social\u0026utm_campaign=Launch_YC_CatchAll\u0026utm_id=catch_all_YC), test CatchAll (for free), and give us feedback (we’ll 5x your credits!).\\\n\\\nShare your use-cases:\\\n[artem@newscatcherapi.com](mailto:artem@newscatcherapi.com) or DM on [https://www.linkedin.com/in/artem-bugara](https://www.linkedin.com/in/artem-bugara/)\n\n![uploaded image](/media/?type=post\u0026id=99812\u0026key=user_uploads/444913/74de4236-76fc-4c2b-8ef4-d02248dd0837)\n\n","slug":"Pxs-catchall-recall-first-web-search-api","created_at":"2026-04-16T15:13:11.193Z","updated_at":"2026-05-25T02:48:07.079Z","total_vote_count":4,"url":"https://www.ycombinator.com/launches/Pxs-catchall-recall-first-web-search-api","share_image_url":"https://www.ycombinator.com/media/?type=post\u0026id=99812\u0026key=user_uploads/444913/74de4236-76fc-4c2b-8ef4-d02248dd0837","company":{"id":26860,"name":"NewsCatcher","slug":"newscatcher","url":"https://newscatcherapi.com/","logo":"https://bookface-images.s3.amazonaws.com/small_logos/4e6df1d33e5151e6b36a83a91e94b8bb2550c930.png","batch":"Summer 2022","industry":"B2B","tags":["Artificial Intelligence","SaaS","Enterprise","Big Data","Enterprise Software"],"search_path":"https://bookface.ycombinator.com/company/26860"}}