Daily top stories

Bicycle

Bicycle

CIECHANOW.SKI
todsacerdoti
1289
186
12h

Infinite Mac: Infinitemac.org

BLOG.PERSISTENT.INFO
mariuz
100
19
10h
UK sets up fake booter sites to muddy DDoS market

UK sets up fake booter sites to muddy DDoS market

KREBSONSECURITY.COM
todsacerdoti
74
46
10h

My4TH – A minimalistic FORTH computer with discrete CPU

MYNOR.ORG
incanus77
116
7
11h

Open Flamingo – open framework to train multimodal LLMs

LAION.AI
mpaepper
4
0
7h
Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

CEREBRAS.NET
asb
341
132
11h
EU Commission doesn't understand what's written in its own chat control bill

EU Commission doesn't understand what's written in its own chat control bill

MULLVAD.NET
rc00
246
57
9h

Launch HN: Metal (YC W23) – Embeddings as a Service

Hey HN! We’re Taylor, James and Sergio – the founders of Metal (https://www.getmetal.io/). You can think of Metal as embeddings as a service. We help developers use embeddings without needing to build out infrastructure, storage, or tooling. Here’s a 2-minute overview: https://www.loom.com/share/39fb6df7fd73469eaf20b37248ceed0f

If you’re unfamiliar with embeddings, they are representations of real world data expressed as a vector, where the position of the vector can be compared to other vectors – thereby deriving meaning from the data. They can be used to create things like semantic search, recommender systems, clustering analysis, classification, and more.

Working at companies like Datadog, Meta, and Spotify, we found it frustrating to build ML apps. Lack of tooling, infrastructure, and proper abstraction made working with ML tedious and slow. To get features out the door we’ve had to build data ingestion pipelines from scratch, manually maintain live customer datasets, build observability to measure drift, manage no-downtime deployments, and the list goes on. It took months to get simple features in front of users and the developer experience was terrible.

OpenAI, Hugging Face and others have brought models to the masses, but the developer experience still needs to be improved. To actually use embeddings, hitting APIs like OpenAI is just one piece of the puzzle. You also need to figure out storage, create indexes, maintain data quality through fine-tuning, manage versions, code operations on top of your data, and create APIs to consume it. All of this friction makes it a pain to ship live applications.

Metal solves these problems by providing an end-to-end platform for embeddings. Here’s how it works:

Data In: You send data to our system via our SDK or API. Data can be text, images, PDFs, or raw embeddings. When data hits our pipeline we preprocess by extracting the text from documents and chunking when necessary. We then generate embeddings using the selected model. If the index has fine-tuning transformation, we transform the embedding into the new vector space so it matches the target data. We then store the embeddings in cold storage for any needed async jobs.

From there we index the embeddings for querying. We use HSNW right now, but are planning to support FLAT indexes as well. We currently index in Redis, but plan to make this configurable and provide more options for datastores.

Data Out: We provide querying endpoints to hit the indexes, finding the ANN. For fine-tuned indexes, we generate embeddings from the base model used and then transform the embedding into the new vector space during the pre-query phase.

Additionally, we provide methods to run clustering jobs on the stored embeddings and visualizations in the UI. We are experimenting with zero-shot classification, by embedding the classes and matching to each embedding in the closest class, allowing us to provide a “classify” method in our SDK. We would love feedback on what other async job types would be useful!

Examples of what users have built so far include embedding product catalogs for improved similarity search, personalized in-app messaging with user behavior clusters, and similarity search on images for content creators.

Metal has a free tier that anyone can use, a developer tier for $20/month, and an enterprise tier with custom pricing. We’re currently building an open source product that will be released soon.

Most importantly, we’re sharing Metal with the HN community because we want to build the best developer experience possible, and the only metric we care about is live apps on prod. We’d love to hear your feedback, experiences with embeddings, and your ideas for how we can improve the product. Looking forward to your comments, thank you!

tlowe11
127
73
14h

Interaction Nets, Combinators, and Calculus – HVM

ZICKLAG.GITHUB.IO
erlend_sh
45
3
11h
Intel 80386 CPU Information

Intel 80386 CPU Information

PCJS.ORG
peter_d_sherman
29
1
1d
TaxyAI: Open-source browser automation with GPT-4

TaxyAI: Open-source browser automation with GPT-4

GITHUB.COM
kcorbitt
70
14
11h
Medieval Arabic surgeon Ibn al Quff's account on surgical pain relief

Medieval Arabic surgeon Ibn al Quff's account on surgical pain relief

JOURNALS.LWW.COM
teleforce
9
0
1d

Apple Music Classical

LEARN.APPLEMUSIC.APPLE
tosh
469
390
20h
Snippyly (YC W22) Is Hiring Front end SWEs to make the web multiplayer

Snippyly (YC W22) Is Hiring Front end SWEs to make the web multiplayer

YCOMBINATOR.COM
rakeshgoyal
1
0
11h

Tectonic – A modern, complete, self-contained Tex engine with Unicode support

TECTONIC-TYPESETTING.GITHUB.IO
Gadiguibou
38
10
1d
Iceland long term visa for remote workers

Iceland long term visa for remote workers

ISLAND.IS
simonebrunozzi
155
140
12h
Show HN: A fully open-source (Apache 2.0)implementation of llama

Show HN: A fully open-source (Apache 2.0)implementation of llama

We believe that AI should be fully open source and part of the collective knowledge.

The original LLaMA code is GPL licensed which means any project using it must also be released under GPL.

This "taints" any other code and prevents meaningful academic and commercial use.

Lit-LLaMA solves that for good.

GITHUB.COM
osurits
74
38
10h

Nuenv: An experimental Nushell builder for Nix packages

DETERMINATE.SYSTEMS
biggestlou
64
17
13h
A non-federated decentralized social protocol based on Git

A non-federated decentralized social protocol based on Git

GITHUB.COM
petar
63
32
13h

The Shaman’s Secrets

ARCHAEOLOGY.ORG
drdee
30
6
22h

Ivy League Prices Are Pushing $90k a Year

MSN.COM
ExMachina73
17
1
8h
Fugue: A unified interface for distributed computing

Fugue: A unified interface for distributed computing

GITHUB.COM
duck
64
11
1d
Android app from China executed 0-day exploit on millions of devices

Android app from China executed 0-day exploit on millions of devices

ARSTECHNICA.COM
LinuxBender
189
71
15h

How to Polis, 101, Part IIb: Archons

ACOUP.BLOG
gostsamo
29
9
14h

Quicker serverless Postgres connections

NEON.TECH
nikita
127
71
13h
Windows 11 KB5023778 update adds promotions to the Start menu

Windows 11 KB5023778 update adds promotions to the Start menu

BLEEPINGCOMPUTER.COM
nazgulsenpai
82
70
9h

Open Sourcing Cody – Sourcegraph's AI-enabled editor assistant

ABOUT.SOURCEGRAPH.COM
michlim
73
27
16h
Amazon starts flagging frequently returned products that you maybe shouldn’t buy

Amazon starts flagging frequently returned products that you maybe shouldn’t buy

THEVERGE.COM
mfiguiere
355
259
13h

Random Fuzzy Thoughts

TIGERBEETLE.COM
todsacerdoti
63
15
16h
Apple introduces Apple Pay Later

Apple introduces Apple Pay Later

APPLE.COM
dm
158
268
12h

You're all caught up

Don't spend all your valuable time here, life is more important Content refreshes every hour, on the hour