Devops Goes Promptastic
summary
Live demos show a UI that calls Gemini and Bard (LangChain + Vertex PALM APIs, text‑bison‑001/chat‑bison‑001) to generate Terraform files via Gemini and run terraform init/plan/apply to create a GitHub repo, a GitHub Actions bot that adds PR comments from GPT‑3.5 turbo during code review, a recorded simulation that scaffolds CI/CD and infra from an uploaded architecture diagram, and a log‑analysis workflow that claims to reduce root‑cause time from ~60 minutes to ~20–30 minutes by correlating Splunk/Kibana data.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 2:58GitHub repos with prompts and Terraform
- 12:00AI code reviews with GitHub bots
- 18:15Infrastructure from diagrams and prompts
- 27:13AI log analysis for incidents
- 31:10Q&A
related talks
41mImproving DevOps workflows with GenAI
31mEnter the Machines: Reducing Friction in DevOps Using AI
01h 15m10x Development: LLMs For the working Programmer
12mDemo Theater: Practical Path to Infrastructure as Code at Scale
20mPremiere
01h 03mCloud Devops
32mFrom Toil To Triumph: Harnessing Agentic AI To Streamline Infrastructure as Code
13mGolden path with GitHub Actions