961
New
The latest blog posts to take off on Hacker News. 1,840 posts.
962
Pgvector Is Now Faster Than Pinecone at 75% Less Cost
Meet pgvectorscale, an open-source PostgreSQL extension that supercharges pgvector’s vector search performance for large-scale vector workloads.
963
How to get the most out of Postgres memory settings
Engineering deep-dives, product updates, and company news from Tembo — the platform for running coding agents across your stack.
964
DuckDB Isn't Just Fast
A whistlestop tour of the cool bits of DuckDB
965
966
Private Cloud Compute: A new frontier for AI privacy in the cloud
Secure and private AI processing in the cloud poses a formidable new challenge. To support advanced features of Apple Intelligence with larger foundation models, we created Private Cloud Compute (PCC), a groundbreaking cloud intelligence system designed specifically for private AI processing. Built with custom Apple…
967
Apple's On-Device and Server Foundation Models
At the 2024 Worldwide Developers Conference, we introduced Apple Intelligence, a personal intelligence system integrated deeply into iOS 18…
968
New GitHub Organization for the Swift Project
Today, we are announcing an exciting development for the Swift programming language: its migration to a dedicated GitHub organization at GitHub.com/swiftlang.
969
Controversial pesticide research all but vanished from a major conference
During hundreds of sessions, a significant topic went missing: scholarly research on one of Corteva’s products – neonicotinoids.
970
Deterioration of local community a major driver of loss of play-based childhood
Kids growing up in close-knit communities where the social ties are thick are more protected from the harms of the phone-based childhood.
971
OpenAI and Apple Announce Partnership
OpenAI and Apple announce partnership to integrate ChatGPT into Apple experiences.
972
Apple Intelligence for iPhone, iPad, and Mac
Apple today introduced Apple Intelligence, the personal intelligence system for iPhone, iPad, and Mac.
973
974
Why Triplebyte Failed
How a once-successful, well-liked company failed, as told from within.
975
The Backbone of Cybersecurity: Hardware Security Modules
Discover how Hardware Security Modules (HSMs) provide robust security solutions for managing cryptographic keys and protecting sensitive data in various industries.
976
Viagra improves brain blood flow and could help to prevent dementia
A new trial conducted by the University of Oxford reveals that sildenafil, commonly known as Viagra, enhances blood flow to the brain and improves the function of brain blood vessels in patients at a heightened risk of vascular dementia.
977
Claude's Character
How character training cultivates traits like curiosity, open-mindedness, and thoughtfulness in Claude using a variant of Constitutional AI.
978
Creating a matchmaker for your multiplayer game
Don't test in production. Test in simulation!
979
The Backrooms of the Internet Archive
Like many bits of Internet Culture, this simple image of an empty series of rooms represents a deep-repressed or recently-remembered memory of a common Internet Legend, or it's just a shot of nothing. If the answer is that it's a shot of nothing, let's get you up to speed. This image floated around message boards
980
Breaking up is hard to do: Chunking in RAG applications
A look at some of the current thinking around chunking data for retrieval-augmented generation (RAG) systems.
981
WARC-GPT: An open-source tool for exploring web archives using AI
The Library Innovation Lab is growing knowledge and community by bringing library principles to technological frontiers.
982
Writing memory safe JIT compilers
How to kill off a top source of browser exploits
983
LedgerStore Supports Trillions of Indexes at Uber
Uber connects the physical and digital worlds to help make movement happen at the tap of a button. Billions of trips, deliveries, and tens of billions of financial transactions across earners, spenders, and merchants are made at Uber every quarter. LedgerStore is an immutable storage solution at Uber that provides…
984
Saving Three Months of Latency with a Single OpenTelemetry Trace
Discover how OpenTelemetry tracing and Coralogix analysis resolve latency issues, proving observability's role in enhancing system performance and reliability.
985
The problem with lying is keeping track of all the lies
Or why clear consistency guarantees are how to stay sane when programming distributed systems.
986
987
Dragonfly: A large vision-language model with multi-resolution zoom
Dragonfly is a vision-language architecture with multi-resolution zoom for fine-grained image understanding, including a medical model beating Med-Gemini.
988
Extracting concepts from GPT-4
Using new techniques for scaling sparse autoencoders, we automatically identified 16 million patterns in GPT-4's computations.
989
Qwen2 LLM Released
GITHUB HUGGING FACE MODELSCOPE DEMO DISCORD Introduction After months of efforts, we are pleased to announce the evolution from Qwen1.5 to Qwen2. This time, we bring to you: Pretrained and instruction-tuned models of 5 sizes, including Qwen2-0.5B, Qwen2-1.5B, Qwen2-7B, Qwen2-57B-A14B, and Qwen2-72B; Having been…
990
Understanding QEMU Devices (2018)
Here are some notes that may help newcomers understand what is actually happening with QEMU devices: