semolex.online

Introduction

Pandas is incredibly powerful but there are always some tricks to make it faster. I tried to generalize some simple yet effective optimizations that can help with performance. Most of them are obvious but often ignored.

Introduction

One of the problems with serverless applications is the lack of shared state between invocations. This is especially true for AWS Lambda functions, which are stateless by design. However, there are situations where you might want to share data between invocations, such as caching frequently accessed data or storing configuration settings. You can use Redis or something similar, but it adds complexity and cost to your architecture. Other options?

Introduction

This script checks all repositories in a directory for uncommitted changes. I noticed that I have a lot of repositories on my machine, and sometimes I forget to commit changes when experimenting with different environments.

This script helps me to check all repositories at once.

Introduction

Python is extremely flexible and powerful, but it’s not always the fastest language. And in many cases it really doesn’t matter. Because raw speed is not the only thing that matters. Python ecosystem has a lot of libraries that are more performant than similar libraries in other languages. But in general, you should know your language and its features to write efficient code. What is even more important - Python gives you a lot of ways to handle the way you do some things. You know - there are multiple ways to create objects in Python.

Introduction

There are many ways to organize your serverless functions. Technically, their name is self-explanatory - they are functions. Provider-agnostic, they are just pieces of code that do something upon execution and can either have a payload or not. However, when you start building a slightly bigger system, you might want to organize them in a more structured way.