#fastapi

#fastapi

Modern ML systems do not succeed because of models alone - they succeed because of the software engineering wrapped around them. Most real-world failures in MLOps come from poor structure, missing configuration, messy environments, unclear APIs, or nonexistent logging, not from bad ML.

What's happening here: Type-safe - Your editor knows repo is a SnippetRepository. Full autocomplete, type checking works. Automatic cleanup - The context manager ensures the database session closes, even if an exception occurs. No global state - Every request gets its own session. No risk of one request interfering with another. Testable - Here's the magic: You override the dependency with an in-memory implementation. Your test doesn't hit the database.

Christopher shares an article titled "The Uselessness of 'Fast' and 'Slow' in Programming." It digs into how the different aspects of software performance span a wide range of orders of magnitude, and how developers can obsess over irrelevant performance details, often losing more time working in suboptimal environments than building what they need with tools they already know. We also discuss an article about why uv is fast, which explains how most of its speed comes from engineering decisions rather than just being written in Rust.

Augment Code's new guide features real frameworks to lead your engineering team to systematic transformation: 30% faster PR velocity, 40% reduction in merge times, and 10x task speed-ups across teams. Learn from CTOs at Drata, Webflow, and Scout's local MCP server lets your AI assistant query real Python telemetry. Call endpoints like get_app_error_groups or get_app_endpoint_traces to surface top errors, latency, and backtraces-no dashboards, Brett came across an unexpected typing error when using Pyrefly on his code.

Christopher Trudeau is back on the show this week, bringing another batch of PyCoder's Weekly articles and projects. We discuss a recent Real Python step-by-step tutorial about programming a FastAPI example application. You practice installing FastAPI, building your first endpoints, adding path and query parameters, and validating endpoints using Pydantic. Christopher covers updates to his Real Python video course about concurrency in Python.

Building a UI in Python usually means choosing between "quick and limited" or "powerful and painful." What if you could write modern, component-based web apps in pure Python and still keep full control? NiceGUI, pronounced "Nice Guy" sits on FastAPI with a Vue/Quasar front end, gives you real components, live updates over websockets, and it's running in production at Zauberzeug, a German robotic company. On this episode, I'm talking with NiceGUI's creators, Rodja Trappe and Falko Schindler, about how it works, where it shines,

Asyncpg is the connector for PostgreSQL and asyncio-flavored Python. Here's how to use it without other libraries on FastAPI and Air projects. Recently I've been on a few projects using PostgreSQL where SQLAlchemy and SQLModel felt like overkill. Instead of using those libraries I leaned on writing SQL queries and running those directly in [asyncpg](https://pypi.org/project/asyncpg/) instead of using an ORM powered by asyncpg. Here's how I got it to work

Using Streamlit allowed me to build a fast and efficient frontend without leaving the comfort of Python, leveraging its data-centric features effectively.