Urgent.News

What's breaking now, across thousands of outlets.

AI

Why ChatGPT Makes You Worse at Coding (and What to Do Instead)

ChatGPT can write better code than most junior developers. It can explain concepts, debug errors, and generate entire functions in seconds. For experienced developers, it is a superpower. For beginners trying to learn how to code, it is a trap. The problem: copying code you do not understand When a beginner asks ChatGPT "write me a function to sort a list in Python," they get perfect code…

ChatGPT can generate high-quality code for junior developers and even provide explanations of complex concepts. However, for those learning to code, relying on ChatGPT can hinder the learning process. When beginners copy code they do not fully understand, they miss out on crucial learning opportunities, such as predicting code output and understanding the reasoning behind each step.

The key to effective learning is an active approach: predicting what the code should do before writing it, writing the solution, and then comparing it to a reference solution. This process allows learners to identify gaps in their knowledge and strengthen their understanding.

For example, when asked to filter a list to include only numbers greater than 5, a learner should predict the best approach (using a for loop, list comprehension, or the filter function), write their solution, and then compare it to a reference solution. If their approach differs from the reference, they should explain why and learn from the difference.

While ChatGPT can be a valuable tool for explaining code, it should not be used to generate code outright. Passive usage, such as asking ChatGPT to write a function without engaging in the learning process, leads to superficial understanding and ultimately hinders progress.

To improve coding skills, learners should adopt this active approach for seven days. Whenever they encounter a piece of code, they should stop and write down their prediction in a comment or elsewhere. After running the code, they should check their prediction and, if it was incorrect, write a brief explanation of why. This practice will help learners catch bugs before running the code and develop a stronger mental model of how code works.

By incorporating this structured prediction loop into their learning routine, learners can leverage ChatGPT as a powerful study aid without falling into the trap of passive code generation. This approach will ultimately lead to a deeper understanding of coding concepts and more effective problem-solving skills.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Why Our LLM Bill Hit $31k Instead of $12k (The 4 Production Traps No Pricing Page Shows)

Our Q2 LLM infrastructure bill came in at $31k against a $12k budget . After reverse-engineering every line item, we found that naive token math (multiplying input + output tokens from a model pricing…

  • "Agentic Loop Tax" causes context compounding, inflating token usage
  • "+32% Production Reality Tax" reduces efficiency through retries and fallbacks
  • "GPU Self-Hosting TCO Mirage" reveals hidden costs of dedicated GPU hosting

Where Are the AI-Generated Killer Apps?

Paul Ford, writing for The New York Times (gift link): For a while, I must admit, it looked as if software developer roles like mine were done for. How could we fight against tireless robots?

More from Saturday 12 September →