PANEL - Unleashing the Future: The Impact of Gen AI on Development Productivity
summary
Panel summarizes hands-on experience with GitHub Copilot, ChatGPT, Midjourney, CodeWhisperer and models like StarCoder, showing concrete wins and real failures: Copilot inserted SQL injection and cross-site scripting when trained on sloppy projects, unit tests and repetitive code get faster with Copilot in IDE, but internal research finds junior devs learn faster without AI. Teams build private models (GPTs trained on docs) that actually speed onboarding, experiments tune JVM flags and generate demo vulnerabilities for security remediation. Concerns: copyright, data leakage, carbon cost, and inflated productivity claims.
AI-written summary · about
chapters (10)
- 0:00Panel introduction
- 6:13Panelist intros
- 11:18Gen AI tool takeaways
- 17:26AI and developer productivity
- 23:11AI beyond coding
- 26:16AI adoption risks
- 31:20AI code copyright issues
- 37:57AI sustainability and future
- 48:07AI’s long-term effects
- 55:10Closing thoughts
related talks
57mAccelerate Your Developer Productivity with AI Embrace the Future Now!
40mBuilding the Future Developer: Trends in Open Source, Education...
51mCode Smarter, Not Harder: AI-Powered Dev Hacks for All
01h 02mDeveloping resilience for the future of work
52mPanel: Productivity gains through AI?!
46mAI Integration for Devs: ChatGPT and Copilot in Your Projects
36mUnleashing Generative AI: Improving Developer Productivity in SRE
48mMastering the AI Shift: Adaptive Strategies for Software Evolution