Panel discussion: AI in code development – hype vs reality
summary
Panel reports that AI is adopted widely but usually brings modest gains: 90% adoption, 72% report only moderate productivity uplift. Practical wins include rapid POCs (hours vs days), test-driven workflows producing cleaner AI-generated code, and reverse-engineering legacy systems plus autogenerated dependency/capability maps; requirements: parsers, knowledge graphs, good engineering foundations. Failures come from overlong prompts, context limits, and agent risks; mitigations: human-in-loop, layered defenses, postmortem vulnerability scans, lock down DB/credential access.
AI-written summary · about
chapters (14)
- 0:00AI in software development
- 2:02AI in daily tasks
- 4:34AI for architecture
- 8:09Rolling out AI in teams
- 12:05AI readiness indicators
- 14:54Responsible AI risks
- 18:50AI and test-driven development
- 21:38Validating AI-assisted code
- 25:05AI in IDEs
- 27:31AI for legacy code
- 32:27AI for large codebases
- 33:53AI adoption guardrails
- 36:44AI for junior devs
- 41:06Closing remarks
related talks
46mPanel Discussion: AI's Impact on Developers
18mHumans + AI: The Future of Software Development Teams
56mPANEL: Unleashing the Future: The Impact of Gen AI on Development Productivity
25mPanel The State of Web Development in 2026
52mPanel: Productivity gains through AI?!
01h 02mPANEL - Unleashing the Future: The Impact of Gen AI on Development Productivity
30mAutomate All the Things Panel Discussion
50mPanel Discussion: What Can Go Wrong With AI in Testing?