TL;DR
Netflix's in-house LLM serving, the milliseconds after you tap pay, and PyPI's new 14-day rule sketch what production AI infrastructure really looks like.
🦄
Articles & Tutorials
× 12 01 Building Agentic Workflows in Python with LangGraph
02 A Data Landscape Guide for Developers
03 Sandboxing an AI Agent
04 Adding python-version to pyvenv.cfg
05 What happens in the milliseconds after you tap pay
06 In-House LLM Serving at Netflix
07 Releases now reject new files after 14 days
08 Deploying Web Apps in 2026
09 PyCon US 2026 Recap
10 Some more things about Django I've been enjoying
11 Django: introducing django-crawl
12 PEP 837: Extensible JSON serialization
🐿️
Projects & Resources
× 12 01 loopgain: Cost Controller for AI Agent Loops
02 protobuf-py: Modern Protocol Buffers for Python
03 ai-agent-book: "In-Depth Understanding of AI Agents: Design Principles and Engineering Practice"
04 LightRAG: Simple and Fast Retrieval-Augmented Generation
05 AI-Infra-Guard: Full-Stack AI Red Teaming Platform
06 outlines: Structured Text Generation
07 WrenAI: GenBI for AI Agents
08 ODS: Osmantic Device Server
09 SenseNova-U1: Native Unified Multimodal Model
10 gs-quant: Python Toolkit for Quantitative Finance
11 headroom: Context Compression for AI Agents
12 Omnigent: Meta-Harness for AI Coding Agents