Show HN: I built a 2B-page search engine, independent of Google/Bing

8 points by Chief_Searcha 330 days ago | 8 comments

Hi HN, For the last 18 months, I've been working solo on building a completely independent search engine from scratch. Today, I'm opening it up for beta testing and would love to get your feedback. The project powers two public sites from the same 2-billion-page index: Searcha.Page: A session-aware search engine that uses a persistent browser key (not a cookie) for better context. Seek.Ninja: A 100% stateless, privacy-first version with no identifiers at all. The entire stack is self-hosted on a single ~$4k bare-metal EPYC server in my laundry room (no cloud, no VC funding). The search pipeline is a hybrid model, using a traditional lexical index for the heavy lifting and lightweight LLMs for specific tasks like query expansion and re-ranking. It's an experiment in capital efficiency and digital sovereignty—proving you don't need Big Tech APIs to compete. I’m looking for feedback on search result relevance, speed, and the clarity of the privacy models. Please try it out and let me know what you think. Links: https://searcha.page https://seek.ninja Thanks, Ryan

ofalkaed 330 days ago |

Really appreciate that when searching for commodities/goods the results are not dominated by amazon/etsy/ebay sellers, very low on multiple results from a single domain in general. Solves my biggest gripe with search engines but have only done a few quick searches with seek.ninja so far.

The one thing that I would like to see is the "More results from <site>," links for sites with a lot of results. :site works so not a deal breaker, just a would be nice.

Chief_Searcha 330 days ago | |

That is a great idea, I will roadmap it. Thank you!

mindctrl-org 330 days ago |

Hi, cool project. A little feedback, as I only spent a few seconds there. The search results need to populate considerably faster. The weird delay, animation, and layout shifting makes it all feel cheap.

The first results were good for the search I did. However, I went back and searched for something else entirely, and I got results similar to the first search. That seems bad, since they were unrelated searches.

Now I can't search anything because it's timing out.

Chief_Searcha 330 days ago | |

Thank you, I will check for errors re the time out. This is just the type of feedback I am looking for. Regarding the overall speed, I have a plan but it won't be easy so it could be a few months to tackle the single biggest issue. Much appreciated!

phillipseamore 330 days ago |

This is great. May I suggest you post a direct link to seek.ninja if this doesn't get the upvotes it deserves today.

Did you crawl yourself or using common crawl?

Chief_Searcha 330 days ago | |

Thank you! Both, I use common crawl as the bulk, but it has long delays so I need to crawl myself too. I'm planning on upgrading my own crawling capabilities but not yet ready since it will require an investment. I also have plans to have a smarter crawl system which I may implement in the next half year.

FerkiHN 330 days ago |

Great work, especially since you did it yourself, you really put your heart and soul into the project and not in vain, it is truly unique and I appreciate it, good luck with further development, you can add local pages (like the most popular services) this will save mobile traffic.

Chief_Searcha 330 days ago | |

Thank you, I appreciate it. For local pages I do have that planned, it will take a bit of time to make a native local index due to many problems such as businesses closing and not reporting it. This is one area that Google has such a strong advantage - user submitted content - and one I would need to find a way to automate. I have various plans and maybe one of them would work. The other option is to outsource it to an API, at least to buy time.