Stenobird

Episode

#471 The ORM pattern of 2026?

Podcast
Python Bytes
Published
Mar 2, 2026
Duration seconds
2363
Processing state
processed
Canonical source
https://pythonbytes.fm/episodes/show/471/the-orm-pattern-of-2026
Audio
https://pythonbytes.fm/episodes/download/471/the-orm-pattern-of-2026.mp3
JSON
/v1/public/podcasts/python-bytes/episodes/471-the-orm-pattern-of-2026
Markdown
/podcast/python-bytes/471-the-orm-pattern-of-2026.md

Actions

  • POST https://stenobird.com/v1/public/podcasts/python-bytes/episodes/471-the-orm-pattern-of-2026/transcription-requests
    Idempotently request low-priority transcript generation for this episode.
  • GET https://stenobird.com/podcast/python-bytes/471-the-orm-pattern-of-2026.md
    Read the agent-friendly Markdown representation of this episode resource.

Summary

An exploration of the 'Raw + Dataclass' architectural pattern as a response to the rise of agentic programming. The episode also reviews new Python utilities for data validation and menu bar management.

Topics

  • Python
  • Software Architecture
  • Dataclasses
  • SQL
  • Agentic Programming
  • Data Validation
  • macOS
  • DevOps

Highlights

  • Main idea: The 'Raw + DC' pattern advocates for using raw SQL alongside Python dataclasses to create a more portable, framework-agnostic data access layer
  • Practical takeaway: Use 'Dataclass Wizard' to easily add JSON, YAML, or environment variable parsing capabilities to standard Python dataclasses via mixins
  • Failure mode: Relying heavily on heavy ORMs like Django's can create tight coupling that makes migrating web frameworks difficult
  • Main idea: Agentic programming requires architectures that are easier for AI agents to navigate and manipulate without breaking complex abstractions
  • Practical takeaway: For macOS users, 'Thaw' serves as a modern, open-source alternative to abandoned menu bar managers like Ice

Chapters

  1. 3:55 The Raw + DC Design Pattern: Discussing a new architectural approach using raw SQL and dataclasses to prepare for an era of AI-driven coding.
  2. 15:45 pytest-check Updates: A brief look at recent updates to the pytest-check library and its usage in testing.
  3. 18:45 Dataclass Wizard: An overview of the Dataclass Wizard library, demonstrating how mixins can add JSON and ENV parsing to dataclasses.
  4. 27:30 macOS Productivity Tools: Reviewing new macOS utilities for app management and the transition from Ice to Thaw for menu bar control.
  5. 30:30 The Future of Software Development: Reflecting on Jevons Paradox and the increasing complexity of software in the age of automation.