Weekly Links (Jul 06 - Jul 12)

This week we dive into fundamentals that define performance and productivity: from caching strategies that sustain every modern application to an honest analysis of where the real bottlenecks in software development actually lie. We also explore valuable perspectives on tools versus code and important reflections on the future of programming careers.

The Complete Guide to Caching

PlanetScale presents a comprehensive guide to caching strategies that go far beyond the basics. The article explores temporal locality (recent data is accessed more) and spatial locality (related data tends to be requested together), replacement policies like LRU and Time-Aware LRU, and layers from CPU to databases. The key is understanding that cache isn't just about speed - it's about balancing "capacity, speed, cost, and durability" at every application layer.

Tools: Code Is All You Need

Armin Ronacher (creator of Flask) questions the hype around MCP (Model Context Protocol) and argues that code generation is more reliable for automation. He demonstrates migrating his blog from reStructuredText to Markdown using LLMs to generate verifiable and reusable Python scripts. His philosophy: "when you don't have an AI... your tool of choice is code." The article shows how transparent and repeatable automation beats complex integrations.

Writing Code Was Never the Bottleneck

A precise analysis of where the real bottlenecks in development lie: code reviews, knowledge transfer, testing, debugging, and coordination. The author argues that "the biggest cost of code is understanding it - not writing it." LLMs can accelerate initial implementation, but don't solve the fundamental collaborative aspects of software engineering. An essential perspective on why productivity goes beyond coding speed.

The Uncertain Future of Coding Careers and Why I'm Still Hopeful

A balanced reflection on AI's impact on programming careers. The author acknowledges legitimate anxieties (layoffs, fear of replacement) but maintains optimism: AI will handle "grunt work", freeing humans for creative problem-solving. Programmers will prosper as "shepherds" of AI tools, providing context and guidance. The central message: "the future of this career isn't about being replaced by a machine - it's about being amplified by one."

🎥 Talks and Presentations

Functional Programming in 40 Minutes - Russ Olsen

Jul 09, 2025 ⸱ 31m 40s

Data: The Land DevOps Forgot - Michael Nygard

Jul 11, 2025 ⸱ 33m 27s

Thinking Deeper in Gemini - Jack Rae

Jul 10, 2025 ⸱ 18m 13s

Next level Kotlin support in Spring Boot 4 - Sébastien Deleuze

Jul 11, 2025 ⸱ 39m 42s

Compose Hot Reload 🔥 is here, stop relaunching your app!

Jul 10, 2025 ⸱ 25m 18s