🗄️ Columnar Storage - Souguir · Scour

SysTools Parquet Splitter Tool for Large File Splitting ⚡Spark

hackster.io·5d

Data Compression and the Meaning of Information 📊Algorithms

algodeck.com·17h·Hacker News

The Hidden Value of CPU-Intensive Compression on Modern Hardware 🏗️System Design

klarasystems.com·1d

Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization ✍️Prompt Engineering

arxiv.org·5h

Fitting Gemma 4 (~52 GB) into 12 GB 🌊Delta Lake

varjosoft.com·2d·Hacker News

We cut Codex's input token cost by 49.5% with a compression gateway (benchmark) 🔧Feature Engineering

edgee.ai·16h·Hacker News

PAX: The Storage Engine Strikes Back 🌊Delta Lake

mydbanotebook.org·4d

Compression, Prediction, and Scale 🔧Feature Engineering

medium.com

·14h

Article: Building Hierarchical Agentic RAG Systems: Multi-Modal Reasoning with Autonomous Error Recovery ❄️Snowflake

infoq.com·1d

Turn Every Sales Follow-Up Into a Closing Plan 🚀SQL Optimization

usepipeline.ai·3h·Hacker News

How Zalando built a unified data foundation for AI and analytics on Databricks ❄️Snowflake

databricks.com·17h

anurag-as/Bonsai: A self-tuning Rust library that transparently switches between spatial index structures at runtime to maintain optimal query performance as your data evolves. 🗂️Data Structures

github.com·1d·Hacker News

How Respan is scaling LLM observability with ClickHouse Cloud 🌊Delta Lake

clickhouse.com·3d

Compression technique makes AI models leaner and faster while they're still learning ✍️Prompt Engineering

techxplore.com·14h

Google TurboQuant Explained: The 6x Memory Compression That Crashed Chip Stocks 🔧Feature Engineering

medium.com·2d

XUASTC's next step: Intra-prediction of weight grids 🔧Feature Engineering

richg42.blogspot.com·15h·richg42.blogspot.com

TurboQuant: How Google’s 3-Bit Compression Makes Local Agent Swarms Possible ⚡Spark

medium.com·6d

Breaking the Memory Wall: TurboQuant KV Cache Quantization on Apple Silicon ✍️Prompt Engineering

pub.towardsai.net

·1d

Geekbench 6.7 adds Intel BOT detection to spoof out 'unrealistic' CPU scores — Benchmark runs with BOT enabled will be marked as invalid ✍️Prompt Engineering

tomshardware.com

·14h

Show HN: I built a local data lake for AI powered data engineering and analytics ⚡Spark

stream-sock-3f5.notion.site·1d·Hacker News

Loading more...