avthar

5 ms·

1.
Postgres vs. Qdrant Vector Database Comparison(timescale.com)
3 points by avthar 1y ago|0 comments
2.
Stratechery Year in Review 2024(stratechery.com)
3 points by avthar 2y ago|0 comments
3.
The Emerging Open-Source AI Stack(timescale.com)
5 points by avthar 2y ago|0 comments
4.
Measuring Moats [pdf](morganstanley.com)
1 points by avthar 2y ago|0 comments
5.
Scaling Document Data Extraction with LLMs and Vector Databases(timescale.com)
12 points by avthar 2y ago|2 comments
6.
by avthar 2y ago
Post-co author here. This is actually something that we are considering implementing in future versions of pgai Vectorizer. You point the vectorizer at database A but tell it to create and store embeddings in database B. You can always do j
7.
by avthar 2y ago
Post co-author here. Really appreciate the feedback. Your point about HNSW being resource intensive is one we've heard. Our team actually built another extension called pgvectorscale [1] which helps scale vector search on Postgres with
8.
by avthar 2y ago
Post co-author here. The point is a little nuanced, so let me explain: You are correct in saying that that you can store embeddings and source data together in many vectordbs. We actually point this out in the post. The main point is that t
9.
by avthar 2y ago
Good q. For most standalone vector search use cases, FAISS or a library like it is good. However, FAISS is not a database. It can store metadata alongside vectors, but it doesn't have things you'd want in your app db like ACID com
10.
by avthar 2y ago
Hey HN! Post co-author here, excited to share our new open-source PostgreSQL tool that re-imagines vector embeddings as database indexes. It's not literally an index but it functions like one to update embeddings as source data gets ad
11.
by avthar 2y ago
While I value the depth in which the blog authors went into, some of these are not problems with pgvector but problems with similarity search itself and would be problems with other vector databases as well. There are also solutions to the
12.
by avthar 2y ago
There's actually quite a few ways to do shared database multi-tenant setups with pgvector. You can have separate out tenants into separate tables, schema, and even logical databases within the same database service itself. Here's
13.
Reflections on using OpenAI o1 / Strawberry for 1 month(oneusefulthing.org)
46 points by avthar 2y ago|4 comments
14.
by avthar 2y ago
See discussion for pgvectorscale in this HN post: https://news.ycombinator.com/item?id=40646276
15.
by avthar 2y ago
Blog co-author here (PM at Timescale). We're excited to release pgvectorscale. Our team built this extension to make PostgreSQL a better database for AI and to challenge the notion that PostgreSQL and pgvector are not performant for ve
16.
Vector Search Queries with Time Filters in Pgvector(timescale.com)
4 points by avthar 2y ago|0 comments
17.
by avthar 3y ago
Timescale recently released Timescale Vector [0] a scalable search index (DiskANN) and efficient time-based vector search, in addition to all capabilities of pgvector and vanilla PostgreSQL. We plan to add the document processing and embedd
18.
by avthar 3y ago
Agreed on pgvector being simple and a great choice for POCs and low scale, especially if you're familiar with Postgres. Our team released something new last week built for folks looking to use PostgreSQL at scale as a vector store [0],
19.
by avthar 3y ago
(Blog author here). Thanks for the question. In this case the index for both DiskANN and pgvector HNSW is small enough to fit in memory on the machine (8GB RAM), so there's no need to touch the SSD. We plan to test on a config where th
20.
by avthar 3y ago
Blog author here. This index is built in PostgreSQL. We make it very clear that this is cloud-only in the post and never mention OSS. While we offer it via the Timescale cloud platform, everything under the hood is still PostgreSQL and work
21.
by avthar 3y ago
Yup 2GB is fine (especially if using quantization), depending on the dataset size of course. Totally fair about cloud-only. Many developers prefer developing on cloud, but some prefer local dev. YMMV. By the way timescale vector offers pgve
22.
“Developers hate being marketed to”(devmarketing.club)
3 points by avthar 3y ago|2 comments
23.
What Is a Technical Product Marketing Manager?(devmarketing.club)
1 points by avthar 3y ago|0 comments
24.
by avthar 4y ago
Source clip: https://twitter.com/adam22/status/1527005564802600960?s=20&t...
25.
Observing Our Observability Tool(timescale.com)
27 points by avthar 4y ago|0 comments
26.
by avthar 5y ago
Timescaler here. Linking a few posts below [0][1] that answer the majority of your very good questions. The posts detail why and how TimescaleDB started and why the founders chose to build a time-series database on PostgreSQL. [0]: https:&
27.
Money and Investing: Book Recommendations(avthar.substack.com)
5 points by avthar 5y ago|0 comments
28.
by avthar 5y ago
Post author here. We made the NFT Starter Kit to help fellow developers make sense of the NFT space using data. We're looking to improve the project in the coming months, so I'd be curious to hear your take and suggestions for imp
29.
by avthar 5y ago
> Timescale and Citus are are open core, holding back features for customers. One clarification. While TimescaleDB is open-core, our community version is source-available and 100% free to use. We do not "hold back features for custo
30.
by avthar 5y ago
Timescaler here. I think you're referring to this comparison of InfluxDB vs TimescaleDB [0]? There's also comparisons of TimescaleDB vs MongoDB[1] and AWS Timestream [2]. [0]: https://blog.timescale.com/blog/t

More ›