The freeCodeCamp Podcast

#215 How to learn programming and CS in the AI hype era – interview with dev and prof Mark Mahoney

13 snips
Apr 10, 2026
Guest
Mark Mahoney
Mark Mahoney, a longtime computer science professor and former developer who built the Playback Press learning platform. He discusses how to learn programming wisely during the AI hype. Short takes on using LLMs for prototyping, avoiding de-skilling, interactive teaching, the value of hands-on projects, and why CS fundamentals still matter now.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
ADVICE

Make LLMs Produce Plans Before Code

  • Ask an LLM for a plan first and iterate on that plan before asking it to write code.
  • Mark iterates on the plan multiple times until it matches his preferences, then has the model generate code.
ADVICE

Step Through LLM Generated Code Yourself

  • Use an LLM as an infinitely patient tutor but actively step through generated code yourself.
  • Mark warns learners must step through code execution to avoid becoming non-resilient if the LLM writes most of the code.
INSIGHT

Unchecked LLM Choices Can Add Technical Debt

  • Heavy LLM reliance risks technical debt and token costs for teams that use them full time.
  • Mark recounts an LLM suggesting a global document object placement that would split state and add debt if unchecked.
Get the Snipd Podcast app to discover more snips from this episode
Get the app