🐿️ ScourBrowse
LoginSign Up
You are offline. Trying to reconnect...
Copied to clipboard
Unable to share or copy to clipboard
📊 Model Serving Economics

GPU Costs, Inference Pricing, Batch Optimization, Resource Efficiency

Disruptive Changes Ahead For Photomasks?
semiengineering.com·15h
🔬Chip Fabrication
Canvas, meet code: Building Figma’s code layers
figma.com·22h
✏️Code Editors
MUVERA: Making multi-vector retrieval as fast as single-vector search
research.google·9h
🎯Qdrant
Visual hallucination detection in large vision-language models via evidential conflict
arxiv.org·18h
🧠LLM Inference
Compbolt: A lib with a hard to misuse API (based on Matt Godbolt)
github.com·13h·
Discuss: Hacker News
🦀Rust Compiler Internals
The more LLMs think, the worse they translate
nuenki.app·10h·
Discuss: Hacker News
🏆LLM Benchmarking
June 25, 2025 Flight Tracking Workshop (4 hour) [Americas / Europe-friendly time]
bellingcat.com·22h
🪄Prompt Engineering
DRIFT: Data Reduction via Informative Feature Transformation- Generalization Begins Before Deep Learning starts
arxiv.org·18h
📊Vector Databases
Comprehensive Comparison of Algorithmic Trading Platforms
jonathankinlay.com·13h·
Discuss: Hacker News
🏗️Infrastructure Economics
A Principled Approach to Randomized Selection under Uncertainty
arxiv.org·18h
📊Statistical Ranking
A large deviation view of \emph{stationarized} fully lifted blirp interpolation
arxiv.org·18h
🧠LLM Inference
You Can Probably Stand to Charge More (2006)
kalzumeus.com·10h·
Discuss: Hacker News
💰Revenue Models
What LLMs Know About Their Users
schneier.com·11h·
Discuss: Hacker News
🪄Prompt Engineering
Black-Box Test Code Fault Localization Driven by Large Language Models and Execution Estimation
arxiv.org·18h
🕯️Candle
Why Dyad?: A Perspective for Modelica Users
juliahub.com·7h·
Discuss: Hacker News
🕯️Candle
in and out, quick appview adventure | futur | WhiteWind blog
whtwnd.com·13h
💧Litestream
MakoGenerate: AI-Powered GPU Kernel Generation in Under 60 Seconds
mako.dev·7h·
Discuss: Hacker News
🖥GPUs
Kumo Surfaces Structured Data Patterns Generative AI Misses
thenewstack.io·8h
🧠LLM Inference
Accelerating hardware development to improve national security and innovation
news.mit.edu·18h·
Discuss: Hacker News
👨‍💻Software development practices
Join me if you can: ClickHouse vs. Databricks & Snowflake - Part 2
clickhouse.com·22h·
Discuss: Hacker News
⚙️Database Internals
Loading...Loading more...
AboutBlogChangelogRoadmap