emschwartz's Likes

T5Gemma: A new collection of encoder-decoder Gemma models  🏗️LLM Infrastructure

Introducing Rudy: A Toolchain for Rust Debuginfo  🦀Rust

OverflowSort: A (possibly) novel high-speed sorting algorithm based on overflow  🔄SIMD Programming

Diversifying search results with Maximum Marginal Relevance  🔗Hybrid Search

Diffusion Elites: surprisingly good, simple and embarrassingly parallel  🆕New AI

The Ancient History of Compression Algorithms  📝Text Compression

Semantic text is all that and a bag of (BBQ) chips! With configurable chunking settings and index options  📄Semantic Chunking

Economics of Claude 3 Opus Inference  📊Model Serving Economics

Announcing Rust 1.88.0  🦀Rust

OpenAI says GPT-5 will unify breakthroughs from different models  🤖AI

Pita Tunnels: A Pain in the Ass for Censors  🌐Pingora

A non-anthropomorphized view of LLMs  🪄Prompt Engineering

RAG chunking isn’t one problem, it’s three  📄Semantic Chunking

A brief history of Internet search  🔎Inverted Index

The messy reality of SIMD (vector) functions  🔄SIMD Programming

Vector Embeddings Quantization with 600% memory footprint reduction  🗜️Vector Compression

Our Fullstack Architecture: Eta, HTMX, and Lit  🧩Web Components

How We Serve Millions of Requests on a Single VM  ⚡Glommio

Felt the need to explain the idea of inbound and outbound RSS feeds and how we take advantage of those in our project  📰RSS Reading Practices

Highly-available, raft-based SQLite  📝SQLite WAL