Skip to content

Essential Reading for Agentic Engineers

Published:
6 min read

Kickstart your transition from vibe coding to agentic engineering.

These resources will help you master the new paradigm of AI-assisted development, where agents become true collaborators that can handle entire codebases and ship production features. Each piece was chosen for its practical, real-world insights. I’ll keep this list updated as the field evolves.

How to Use Claude Code Effectively

Read the article by Philipp Spiess (@philippspiess) • June 2025 • 14 min

Philipp shares practical workflow tips for maximizing Claude Code’s effectiveness:

Claude Code has fundamentally changed how I approach many programming tasks… The key is finding the right balance between automation and human oversight.

Agentic Coding: The Future of Software Development with Agents

Watch the video by Armin Ronacher (@mitsuhiko) • June 2025 • 37 min

Armin explores agentic coding as a transformative paradigm where AI agents become true collaborators rather than mere autocomplete tools, describing it as “catnip for programmers” that draws developers into addictive, energizing sessions. Claude Code emerges as the leading tool, with its terminal-based interface enabling powerful agent chaining and remote workflows that editor integrations can’t match.

The early adopter programmers are just the beginning of a broader wave of agentic AI use.

I can’t sleep gud anymore - A Practical Guide to Agentic Computering

Watch the video by Mario Zechner (@badlogicgames) • June 2025 • 85 min

Mario shares his hard-won patterns for using Claude Code effectively on real projects, demonstrating workflows from ad-hoc scripting to complex cross-language porting live, revealing how to maintain control while leveraging AI’s speed.

It’s not about the AI being smart - it’s about being smart about the AI.

Claude Code Best Practices from Anthropic

Read the article by Anthropic Engineering • April 2025 • 15 min

Anthropic’s engineering team shares comprehensive best practices for maximizing Claude Code’s effectiveness:

Claude performs best when it has a clear target to iterate against: a visual mock, a test case, or another kind of output.

Building a macOS App Entirely with Claude Code

Read the article by Indragie Karunaratne (@indragie) • July 2025 • 19 min

Indragie built and shipped Context, a macOS app for debugging MCP servers, with Claude Code writing ~19,000 of the 20,000 lines of code. Key insights:

This represents a fundamental shift in how developers can work: AI agents aren’t just assistants but capable collaborators that can handle entire codebases.

It’s like I found an extra 5 hours every day, and all it cost me was $200 a month.

A Cautionary Perspective on AI Coding

Read the article by Alberto Fortin (@a7fort) • May 2025 • 8 min

Alberto offers a sobering counterpoint to AI coding enthusiasm after discovering fundamental structural issues in his AI-generated codebase. His key insights:

I’m not asking it to write new things from scratch… I’m the senior dev. The LLM is the assistant.

The Evolution of Software: From 1.0 to 3.0

Watch the video by Andrej Karpathy (@karpathy) • June 2025 • 40 min

Andrej presents a framework for understanding software evolution through three paradigms: Software 1.0 (traditional code), Software 2.0 (neural networks), and Software 3.0 (LLMs programmable via natural language). LLMs are becoming the new operating systems where English is the programming language, fundamentally changing how we interact with computers.

We are at a historic moment where software is fundamentally changing… Being fluent in all three paradigms will be valuable for future engineers.

New posts, shipping stories, and nerdy links straight to your inbox.

2× per month, pure signal, zero fluff.


Edit on GitHub