Panel Discussion: Developing with AI
summary
Five product leaders compare concrete wins and failures: companies ship value by using a good-enough model plus cleaned data and evaluation pipelines, not waiting for the latest model, and expect months not days to production. Typical early wins are lead qualification, meeting briefings and CSV extraction; harder wins include predictive maintenance and drug co‑scientist research that stalls at productization. Successful rollouts use human-in-the-loop checkpoints, start with narrow pilots (many orgs quit after ~6 months), scale over ~12 months, and often target Python/JS/TS stacks.
AI-written summary · about
chapters (19)
- 0:00Panel introduction
- 1:40Common AI project pitfalls
- 5:14Human-AI collaboration challenges
- 6:37Where AI delivers real value
- 9:05Use cases for self-healing apps
- 10:28Starting with simple AI workflows
- 12:14Balancing complexity and value
- 13:58AI co-scientist project journey
- 16:51Factors for AI success
- 18:59AI maturity prerequisites
- 21:47Adoption mindset and culture
- 23:10Software compatibility for AI
- 24:12Early adopters and champions
- 25:55User needs vs product vision
- 28:27Building for enterprise adoption
- 30:11Technology adoption gap
- 31:35Solving real business problems
- 32:37Big wow moments in AI
- 35:28No-code AI application building
related talks
01h 09mWorkshop: Product development for identifying and leveraging ML opportunities
41mPanel discussion: AI in code development – hype vs reality
59mPanel Discussion: Strategies for Scaling AI in Quality Engineering
27mBuilding AI applications: Successes and Lessons
23mDelivering business success with AI
40mData Products and AI: A Conversation on Success Factors and Hurdles
53mThe Human Side of AI: Unlocking Opportunity in the New Bottlenecks of Software Development
33mEngineering AI Applications: A Conversation with AI Leaders