I don't like NumPy

The author, once a fan of NumPy, now criticizes its complexity and opacity when working with high-dimensional arrays, arguing that common operations often become unreadable and error-prone due to confusing broadcasting, indexing, and function conventions. While NumPy excels at simple cases, the post contends that its design choices—especially around implicit broadcasting and lack of expl...

https://dynomight.net/numpy/

LiveSplat

Live Gaussian Splatting for RGBD Camera Streams.

https://github.com/axbycc/LiveSplat

Juvio

UV kernel for Jupyter.

https://github.com/OKUA1/juvio

pyfuze

pyfuze makes your Python project run anywhere.

https://github.com/TanixLu/pyfuze

Turning Data into Insight

The article demonstrates how to build a flexible, modern data lakehouse architecture using open-source tools like MinIO, Apache Iceberg, Airflow, dbt, Spark, Pandera, and Superset. By integrating these technologies with Docker for easy deployment, it shows how to orchestrate robust data pipelines, ensure data quality, and enable scalable analytics from raw ingestion to interactive dashboards.

https://towardsdev.com/turning-data-into-insight-flexible-lakehouse-with-minio-iceberg-airflow-dbt-spark-pandera-409d036e5542

Machine Learning Prototyping with DuckDB and scikit-learn

In this post, we prototype a machine learning workflow using DuckDB for data handling and scikit-learn for modeling.

https://duckdb.org/2025/05/16/scikit-learn-duckdb.html

nlweb

Building conversational interfaces for websites is hard. NLWeb seeks to make it easy for websites to do this. And since NLWeb natively speaks MCP, the same natural language APIs can be used both by humans and agents.

https://github.com/microsoft/nlweb

Flowfile

Flowfile is a visual ETL tool combining drag-and-drop workflows with the speed of Polars dataframes. Build and analyze data pipelines without code. Perfect for analysts and engineers needing fast, intuitive data processing. Designed to run locally or deploy to production environments.

https://github.com/Edwardvaneechoud/Flowfile/

Datatune

Perform transformations on your data with natural language using LLMs

https://github.com/vitalops/datatune

Unravelling t-strings

PEP 750 introduced t-strings for Python 3.14. In fact, they are so new that as of Python 3.14.0b1 there still isn't any documentation yet for t-strings. As such, this blog post will hopefully help explain what exactly t-strings are and what you might use them for by unravelling the syntax and briefly talking about potential uses for t-strings.

https://snarky.ca/unravelling-t-strings/

Ruff - A Fast Linter & Formatter to Replace Multiple Tools and Improve Code Quality

This video is a hands-on tutorial showing how to use Ruff, a super-fast Python linter and formatter written in Rust that consolidates tools like Flake8, Black, and isort into a single, efficient solution. The guide covers installing Ruff, running it from the command line, configuring it for projects, and integrating it with VS Code to improve code quality and developer workflow.

https://www.youtube.com/watch?v=828S-DMQog8

ii-agent

A new open-source framework to build and deploy intelligent agents.

https://github.com/Intelligent-Internet/ii-agent

Beyond Query Optimization

Lyft engineers detail how they improved the scalability and reliability of their Aurora Postgres databases by implementing connection pooling with SQLAlchemy and Amazon RDS Proxy. The article explains the challenges of managing database connections in high-traffic environments and describes how these solutions reduced connection limits, improved application stability, and optimized resou...

https://eng.lyft.com/beyond-query-optimization-aurora-postgres-connection-pooling-with-sqlalchemy-rdsproxy-200db7f562d7

A leap year check in three instructions

The article explores how to check if a year is a leap year using just three CPU instructions, leveraging clever bit manipulation and "magic numbers" to optimize the standard algorithm. By reverse-engineering and brute-forcing constants, the author demonstrates a branchless, highly efficient leap year check for years up to 102,499, illustrating both the mathematical tricks and practical l...

https://hueffner.de/falk/blog/a-leap-year-check-in-three-instructions.html