TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: NNext.net – A Firebase-like managed vector storage for ML applications

37 pointsby wakahiuabout 4 years ago
Hi HN. Peter here. As a machine learning engineer, I mostly think in terms of feature vectors, embeddings, and matrices. One of the most useful byproducts of deep neural networks is embeddings because they allow us to represent high-dimensional data in terms of lower-dimensional latent vectors. These feature vectors can be used for downstream applications like similarly search, recommendation systems and near duplicate detection.<p>As an ML engineer, I was frustrated by the lack of a datastore in which vectors are first-class citizens. As a result, most ML engineers, including myself, end up using awkward workarounds to store vectors such as arrays in SQL&#x2F;NoSQL databases, stringifying vectors and storing them as text in in-memory-based caching systems such as Redis ETC. Furthermore, these systems don&#x27;t allow for vector-based query operations such as nearest neighbor search. Consequently, engineers have to deploy additional approximate nearest neighbor search systems such as Facebook&#x27;s FAISS or Spotify&#x27;s ANNOY. These systems, while nifty and fast, are difficult to install and are costly to maintain. To address these issues, I built NNext, a managed vector datastore in which vectors are first class citizens. NNext, allows you to store vectors along with any json-blob metadata. Furthermore, NNext comes with a fast approximate nearest-neighbor (ANN) search capability.<p>I would love to get feedback on your experience as Data Scientist or ML engineer storing feature vectors and ANN systems. Please shoot me an email at [redacted].

9 comments

vladfabout 4 years ago
I can&#x27;t wait to see the vector DBs we&#x27;re going to have 5 years from now when everyone needs to serve their embeddings. It&#x27;s clearly early but frothy right now.<p>Also check out this similar co I ran into: <a href="https:&#x2F;&#x2F;www.pinecone.io&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.pinecone.io&#x2F;</a> (the CEO, EL, has some classic sketches and feature hashing papers).
评论 #26334739 未加载
评论 #26325129 未加载
loxiasabout 4 years ago
Every time I&#x27;ve needed something like this (3 times in my career, though none in the past 5 years) I&#x27;ve ended up having to implement it from scratch.<p>Depending on the topology of the data, different metrics are used and you can make different performance tradeoffs. And differing business use cases make it hard to find a one-size-fits-all. As I&#x27;m sure you&#x27;re aware, there are dozens of fast ANN approaches, and the code implementing the ANN often is the tightest loop, making hard to be pluggable and performant.<p>TileDB is also quite interesting.<p>Best of luck though, I agree with you there&#x27;s a &quot;missing product&quot; to be invented, I&#x27;m not sure this is exactly it, but I don&#x27;t think I know any better.
malux85about 4 years ago
I love this idea, but there&#x27;s no way a managed service like this is going to work, for two reasons: Performance and compliance.<p>Compliance is pretty obvious and even if you don&#x27;t store the feature dictionaries and only the vectors, that&#x27;s a hard conversation to have with the compliance team. I GET IT, without the feature dictionaries the vectors are useless, I KNOW this is how it works from a technical point of view, but the compliance team still won&#x27;t sign off on it. That&#x27;s just the way of the world.<p>And much much more important is performance, uncompressed high dimensional features are huge and even if you use run-length-encoding or sparse vector storage in the protocol plus some lossless compression, I have trouble keeping the GPU fed from disk, let alone over the network, it&#x27;s going to be multiple orders of magnitude too slow. If the benefits is not claimed to be streaming but are fast vector similarity, keep in mind I can do cosine similarity on literally millions of vectors a second on a single CPU core using vanilla numpy, this was fast enough for me to implement realtime face recognition vector search for Dubai airport, so pretty high scale operations.<p>I&#x27;ve love a self hosted version of this, optimised for I&#x2F;O throughput to the GPU. That would be great.
评论 #26325586 未加载
thegeekbinabout 4 years ago
Not on topic for the actual product... but <i>please</i> remove that terrible scrolling behaviour. I hate when websites do this, it makes the user experience (imo) much worse to just &quot;jump&quot; sections if I try to scroll a tiny bit.
评论 #26329028 未加载
KhoomeiKabout 4 years ago
What I&#x27;ve done for most of my research projects is just pickle a PyTorch dataset object that contains all my embeddings. The pkl file can then just be uploaded anywhere and becomes plug and play with any Torch model.<p>What advantages would this bring for a user like me? I guess it might make more sense for people working closer to production?
评论 #26325015 未加载
评论 #26325067 未加载
phenkdoabout 4 years ago
How&#x27;s this different from Milvus?
评论 #26325000 未加载
villgaxabout 4 years ago
Recently learnt about FAISS &amp; Annoy by Spotify, any comparisons done?
评论 #26325878 未加载
wanwanisgoodabout 4 years ago
Why not use an opensource vector search engine&#x2F;database like Milvus? <a href="https:&#x2F;&#x2F;github.com&#x2F;milvus-io&#x2F;milvus" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;milvus-io&#x2F;milvus</a> It&#x27;s probably one of the most popular solution now.
shanghaikidabout 4 years ago
Milvus is really easy to use, and it&#x27;s oss.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;milvus-io&#x2F;milvus" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;milvus-io&#x2F;milvus</a>