AI missteps as stepping stones
summary
Notes give concrete examples: LLM coding assistants swap between PySpark and Pandas causing type errors and broken notebooks when prompts are imprecise; naive requests produce inconsistent APIs. Remedies are precise prompts, small incremental tests, and feeding academic pseudo‑code plus sample data into Claude or Databricks assistants to get a few‑hundred‑line Python implementation that requires hours of debugging but saves days. A Lego mosaic (dog image) initially uses non‑buyable colors and wrong sizing and after nudging yields buildable bricks, which doubles productivity.
AI-written summary · about
chapters (8)
- 0:00Introduction to AI in data
- 1:39Learning new skills
- 3:22Overcoming coding challenges
- 5:21Avoiding over-reliance
- 6:47Collaborating with LLMs
- 10:01The Lego mosaic case study
- 14:04Best practices and conclusion
- 15:25Q&A session
related talks
39mBack to Basics: Cutting Through the AI-Generated Code Noise
39mAI-assisted development: Common pitfalls and smarter practices
45mThe AI Revolution in Software Development: From Code Writing to Quality Assurance
42mAI First Learning
42mBuild Without Limits - Lessons Learned Working with Data and AI Coding Assistants
40mCode Rot at Machine Speed: Preventing AI-Induced Technical Debt
48mMastering the AI Shift: Adaptive Strategies for Software Evolution
09mLessons Learned in AI Engineering