February 3, 2026
Interns, not autopilot
Ask HN: Anyone else struggle with how to learn coding in the AI era?
AI can’t be your brain—learn the hard stuff, say coders
TLDR: A teacher says AI is just turbo documentation, not a replacement for learning algorithms, design, and curiosity. Commenters clash: some call AI a “teenage intern” and demand manual mastery, others recommend games and practice, and many say it’s simply too early to know the best AI-era learning path.
A university teacher on Hacker News dropped a calm take: AI is basically super-powered documentation, great for explanations but not a shortcut to real skills. You still need algorithms, design, and an understanding of different programming styles—and yes, curiosity—plus old-school resources like Coursera. The community promptly turned it into a spicy group chat.
Hardliners came out swinging. One popular refrain: “Treat AI like a teenage intern.” Use it for things you already understand, not for learning from scratch. Another bold claim: if you can’t code without an AI, you don’t actually know how to code. Cue groans, cheers, and a thousand eye-roll emojis. Meanwhile, the self-taught crew echoed caution: AI learning feels like copy-pasting from forums, so “understand every line before you hit commit”—that’s nerd-speak for saving your changes.
Then came the curveballs. One commenter pitched Zachtronics puzzle games as a brain gym, with the reminder that if programming isn’t fun, there are better hobbies. And the vibe check: we’re too early to know how the “AI learner” grows up. Jokes flew about LLMs (large language models) being “interns that hallucinate” and needing coffee breaks. The consensus? Use the bots, but don’t let them be your brain. Read code. Build stuff. Stay curious.
Key Points
- •AI tools are positioned as advanced interactive documentation that can provide factual information, code, and explanations.
- •The author argues AI alone is insufficient for learning to code, similar to how documentation alone was previously inadequate.
- •Essential skills to develop include algorithmic proficiency (e.g., sorting and graph algorithms), high-level design/architecture, and understanding language paradigms.
- •Learners should study established codebases and learn both strongly-typed and dynamic languages to grasp trade-offs and maintainability.
- •Books and online platforms like Coursera remain effective resources, and curiosity is highlighted as the most important factor in learning.