The future of development: Are our jobs getting harder or easier?
summary
Historical arc from punch cards to AI yields concrete tradeoffs: Quarkus hides boilerplate (convenience logger) but required a manual workaround for a JVM instanceof bug found by reading 20,000 lines of assembler; benchmarks show higher-level frameworks can outperform hand‑written code; LLM assistance makes juniors ~26% more productive while producing wrong answers over half the time and contributing ~40% of code with only 30% acceptance, and YAML surprises (e.g. "no" → false) increase cognitive load.
AI-written summary · about
chapters (14)
- 0:00Developer experience changes
- 1:13Punch cards to AI assistance
- 6:32Higher-level abstractions rise
- 8:16Gatekeeping and skill myth
- 12:28Rust’s performance trade-offs
- 14:33Higher-level languages speed
- 17:45JVM performance tuning
- 23:05When abstraction makes things harder not easier
- 24:47Frameworks and reinventing wheel
- 29:12AI coding assistants
- 37:33AI not productivity silver bullet
- 40:18Cognitive load of development
- 42:06AI taking jobs?
- 45:16Summary: Abstraction and shift down
related talks
01h 02mFuture of developers in the age of AI
42mA Developer's Search for Meaning: Thriving as AI Transforms Our World
04mHow it felt like
01h 01mWill AI Replace My Job? A Town Hall for Software Developers
56mFrom Human Computers to AI - the history and future of software developers
29mBeyond Code: The Role of Developers in the Age of AI
15mThe AI Shift: Rethinking the Developer’s Role in a Machine-Driven World
47mSeven Habits of Highly Effective AI Java Coding