Building a web search engine from scratch with 3B neural embeddings

Web and SaaS

search engine

SHOW HN2025

699points
118comments
#39of 3,837 in 2025
Top 2%of 2025 by points

How HN took it

MixedNo clear lean

29 positive, 12 neutral or asking, 2 mixed, 11 critical, out of 54 comments about the project. Thread score +0.41.

Idea: 7 for, 2 againstExecution: 1 for, 8 againstDocs: 7 for, 0 againstPricing: 5 for, 0 againstDesign: 1 for, 0 against

The thread, comment by comment

Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.

positive neutral or mixed critical about another commenter or something else the maker

What HN said

What people liked

  • Excellent, detailed write-up

    From 3 comments
    • Thank you for sharing! This is one of the coolest articles I have seen in a while on HN. I did some searches and I think the search results looked very useful so far. I…
      de6u99er on HN
    • This must be the best technical article I read on HN in months!
      gbrindisi on HN
    • Brilliant write up - learnt a lot.
      siquick on HN
  • Impressive solo effort producing results that rival major engines

    From 3 comments
    • This is incredibly, incredibly cool. Creating a search engine that beats Google in quality in just 2 months and less than a thousand dollars. Really great idea about the…
      1gn15 on HN
    • This wasn't even in the realm of what I thought is possible for a single person to do. Incredible work! It doesn't seem that far in diatance from a commercial search engine?…
      dcreater on HN
    • Incredibly cool. What a write-up. What an engineer.
      tmelm on HN

Objections

  • Vector-only search misses exact keyword matches

    From 1 comment
    • A vector-only search engine will fail for a lot of common use cases where the keywords do matter. I tried searching for `garbanzo bean stew` and got totally irrelevant bean…
      jvia on HN
  • Ranking favours meta-pages over specialist sources

    From 1 comment
    • Very cool project! Just out of interest, I sent a query I've had difficulties getting good results for with major engines: "what are some good options for high-resolution…
      poly2it on HN
  • Embeddings are open to SEO and prompt-injection style manipulation

    From 1 comment
    • The author claims "It should be far less susceptible to keyword spam and SEO tactics." however anyone with a cursory knowledge of the limitations of embeddings/LLM's knows the…
      zipy124 on HN

Requests

  • Open source the code

    From 2 comments
    • This is really really cool. I had earlier wanted to entirely run my searches on it and though that seems possible, I feel like it would be sadly a little bit more waste of time…
      Imustaskforhelp on HN
    • Thank you for sharing! This is one of the coolest articles I have seen in a while on HN. I did some searches and I think the search results looked very useful so far. I…
      de6u99er on HN
  • Add Common Crawl or other open crawl data

    From 3 comments
    • At the end, the author thinks about adding Common Crawl data. Our ranking information, generated from our web graph, would probably be a big help in picking which pages to crawl.…
      ccgreg on HN
    • This is really really cool. I had earlier wanted to entirely run my searches on it and though that seems possible, I feel like it would be sadly a little bit more waste of time…
      Imustaskforhelp on HN
    • Not sure where you are based, but if you were in the EU and had no commercial intentions, you might want to consider adding the crawls from OpenWebSearch.eu, an EU-funded…
      jll29 on HN

Questions

  • Does the demo still work, given CORS errors and no results reported

    From 3 comments
    • Getting a CORS error from the API - is the demo at https://search.wilsonl.in/ working for anyone else?
      Flux159 on HN
    • I couldn't get the search working (there was some cors error) . But what a feat and writeup. Wonder Stuck!
      rkunnamp on HN
    • Is search.wilconl.in still up? I do not get any search result.
      iamcreasy on HN
  • How much did it cost

    From 1 comment

Alternatives named

  • Google

    From 2 comments
    • Tried this search: What is an sbert embedding? Google still gave me a better result: https://towardsdatascience.com/sbert-deb3d4aef8a4/ Nevertheless this project looks great and…
      rkagerer on HN
    • This wasn't even in the realm of what I thought is possible for a single person to do. Incredible work! It doesn't seem that far in diatance from a commercial search engine?…
      dcreater on HN
  • Kagi

    From 1 comment
    • please, I want to pay for this. 10x better than Kagi which I stopped paying for
      profstasiak on HN
  • YaCy

    From 1 comment
    • This is incredibly, incredibly cool. Creating a search engine that beats Google in quality in just 2 months and less than a thousand dollars. Really great idea about the…
      1gn15 on HN

Category, keywords, summaries and labels on this page were made by AI (Thread Reader v2) and not checked by hand. Each point links to the comments it came from.

Add a badge to your READMEShown on HackerShow
[![Shown on HackerShow](https://hackershow.org/badge.svg)](https://hackershow.org/show/44878151)