VectorVFS
Data
vector search
SHOW HN2025
How HN took it
7 positive, 16 neutral or asking, 3 mixed, 10 critical, out of 36 comments about the project. Thread score +0.03.
Idea: 6 for, 2 againstPerformance: 0 for, 6 againstDesign: 0 for, 2 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
Clever idea of storing embeddings in inodes
From 2 comments
Fun idea storing embeddings in inodes! Very clever! I want to point out that this isn’t suitable for any kind of actual things you’d use a vector database for. There’s no notion…
malcolmgreaves on HNthis is actually a great idea
natas on HN
Nostalgia for BeFS and WinFS style metadata search
From 1 comment
This immediately made me nostalgic for BeOS's BeFS or Windows Longhorn's WinFS database filesystems, and how this kind of thing would have fit them perfect. So much cool stuff…
pseudosavant on HN
Objections
Search is linear over all files, not suited to real vector database use
From 2 comments
Fun idea storing embeddings in inodes! Very clever! I want to point out that this isn’t suitable for any kind of actual things you’d use a vector database for. There’s no notion…
malcolmgreaves on HNIf there is no indexing, how will your search time not increase linear or worse with the number of files?
PeterStuer on HN
Zero-overhead indexing claim is misleading
From 1 comment
> Zero-overhead indexing Embeddings are stored as extended attributes (xattrs) on each file, eliminating the need for external index files or services. Ain't no such thing as…
yencabulator on HN
Docs lack GPU backend and uninstall details
From 1 comment
Great idea indeed. The documentation needs a bit more information to be useful. What GPU backends are supported for example? How do I delete the embedding information after I…
anotherpaul on HN
Requests
Support for macOS
From 1 comment
I wonder if I could use this locally on my macbook. The finder applications built-in search is kinda meh.
adenta on HN
Explain how to delete embeddings
From 1 comment
Great idea indeed. The documentation needs a bit more information to be useful. What GPU backends are supported for example? How do I delete the embedding information after I…
anotherpaul on HN
Add an embedded vector index such as Weaviate
From 1 comment
Might be interesting to add an optional embedded Weaviate [1] with a flat-index [2] to the project. It wouldn't use external services and is fully disk-based. Would allow you to…
thirdtrigger on HN
Questions
Is the embedding for the whole file or per chunk?
From 1 comment
is the embedding for the whole file? or each 1024/512 byte chunk?
asadawadia on HN
How does it compare with LanceDB?
From 1 comment
Files-as-vector stores is LanceDB's value proposition. How do you compare in performance, etc.?
esafak on HN
How can users debug why a file surfaced?
From 1 comment
If VectorVFS obscures retrieval logic behind opaque embeddings, how do users debug why a file surfaced—or worse, why one didn’t?
b0a04gl on HN
Alternatives named
LanceDB
From 1 comment
Files-as-vector stores is LanceDB's value proposition. How do you compare in performance, etc.?
esafak on HN
Finder search
From 1 comment
I wonder if I could use this locally on my macbook. The finder applications built-in search is kinda meh.
adenta on HN
BeFS and WinFS
From 1 comment
This immediately made me nostalgic for BeOS's BeFS or Windows Longhorn's WinFS database filesystems, and how this kind of thing would have fit them perfect. So much cool stuff…
pseudosavant on HN