Things Your AI Team Would Like Their DevOps Teams to Know
summary
Predicts a nine month project that yields nothing because of poor and missing data and shows why nine months of work with no signal matters; names tools actually used like Azure AKS, Azure Machine Learning, GitHub Copilot, Microsoft Fabric, Snowflake and RavenDB and reports messy realities: 14-person teams with different Python versions, models hidden on laptops, model poisoning and fake ChatGPT references; cites DORA findings — 25% adoption but 1.5% less throughput — and urges MLOps, dataset versioning and testing instead of buzzword architectures.
AI-written summary · about
chapters (11)
- 0:00AI and DevOps collaboration
- 3:06Misaligned expectations
- 8:44Tool adoption and unmanaged technology risks
- 13:21AI teams and uncertainty
- 17:57Business priorities alignment
- 22:11Integration challenges
- 26:05Cultural resistance issues
- 31:24Change management
- 36:30AI ethics and governance
- 41:09AI testing and monitoring
- 44:21Q&A
related talks
39mSimplify, speed and improve development with DevOps
31mWant to adopt AI in your business: Good luck!
56mData, AI, Hype and Fear - Recipe for Disaster?
37mThe AI Factor: The Blueprint for Intelligent Deployment
50mRealities & Misconceptions about Enterprise Applications of AI/ML and Data
52mDevOps for AI in the Microsoft World
37mPanel Discussion: Developing with AI
39mDevOps in the Age of AI: Human Powered Evolution