Top all-time · #8304
Wilson Lin
Personal blog · blog.wilsonl.inprogrammingdatasystems
Wilson Lin's personal blog at blog.wilsonl.in covers programming, data and systems. Since 2024, 9 different Hacker News readers have linked it 14 times, most often “Show HN: Building a web search engine from scratch with 3B neural embeddings”. It is still writing: the last post was on 2025-11-03, and you can follow it by feed.
Too few sampled comments (7) to say how often HN recommends it.?
Why HN links it
HN readers link Wilson Lin's posts on building search engines, including a DOM-graph write-up, HNSW vector search, and an archiving analysis. Commenters cite them as sources and as learning material.
Summarised by Claude from a sample of 7 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Blog for Wilson Lin
Most-linked posts?
- Show HN: Building a web search engine from scratch with 3B neural embeddings5 people · 6 mentions · first linked 2025-08-12
- Building blobd: single-machine object store with sub-ms reads and 15 GB/s upload2 people · 2 mentions · first linked 2025-11-03
- Show HN: Exploring HN by mapping and analyzing 40M posts and comments for fun2 people · 2 mentions · first linked 2024-05-09
What Hacker News says?
Not exactly what you’re looking for but Wilson Lin’s search engine creates a graph from the DOM for context. Here’s his write up: https://blog.wilsonl.in/search-engine/
Comment linking /search-engine/
on “So you wanna build a local RAG?”Submitted Show HN: Building a web search engine from scratch with 3B neural embeddings as a story · 699 points
HN discussionI absolutely devoured Wilson Lins articles recently .. they are very high quality and informative for any amateur interested in search engines and LLMs! - https://blog.wilsonl.in/search-engine/
Comment linking /search-engine/
on “Guy running a Google rival from his laundry room”Submitted Show HN: Exploring HN by mapping and analyzing 40M posts and comments for fun as a story · 520 points
HN discussionIt claims I reached the article limit. The last time I saw a fastcompany link must have been a decade ago! I was nostalgically looking forward to read another article of theirs. Alas... https://archive.is/HA7y4 Some bits and pieces: > his new search engine, the robust Search-a-Page < https://searcha.page >, which has a privacy-focused variant called Seek Ninja < https://seek.ninja > > The secret to making it all hap…
Comment linking /search-engine/
on “Guy running a Google rival from his laundry room”Submitted Mapping Hacker News to find who knows what in the HN community as a story · 379 points
HN discussionTwo resources that helped me understand HNSWs: https://blog.wilsonl.in/graph-vector-search/ https://github.com/brtholomy/hnsw
Comment linking /graph-vector-search/
on “Scaling HNSWs”Submitted Building blobd: single-machine object store with sub-ms reads and 15 GB/s upload as a story · 65 points
HN discussionSomeone did that and found that only 5% of submitted URLs (200k/4M) were dead: https://blog.wilsonl.in/hackerverse/ Submitted 3 months ago: https://news.ycombinator.com/item?id=40307519
Comment linking /hackerverse/
on “To preserve their work journalists take archiving into their own hands”Submitted Show HN: 40M embeddings to find who knows what on HN as a story · 33 points
HN discussionI'd have to assume it's a blob store for their search engine (or similar) project: https://blog.wilsonl.in/search-engine/
Comment linking /search-engine/
on “Building blobd: single-machine object store with sub-ms reads and 15 GB/s upload”Submitted Building blobd: single-machine object store with sub-millisecond reads as a story · 4 points
HN discussionPartially, the problem here is that the demand for classic search has been reduced quite a bit with LLM-based search finding wide usage amongst common people. You could single out Google for it, as the DoJ and some other entities are doing, but even in that case someone else would take that place with the same dynamics, such as OpenAI or Perplexity. Also, while building search is complex, it’s also not as unfathomab…
Comment linking /search-engine/
on “AI Mode in Search gets new agentic features and expands globally”Submitted Show HN: Searching 1B embeddings from disk in 15 ms on a single machine as a story · 3 points
HN discussion
