Closing panel: The future of platform engineering and what to build next
summary
Platform engineering must solve concrete problems: surviving a withdrawal event like Fable 5 being live for ~3 days then pulled; containing agentic chaos so autonomous agents don’t leak data or run endless, costly loops (examples include 24‑hour agent runs burning impractical compute); and building optionality with multiple LLMs, harnesses that route cheap models for simple tasks, enforced policy as code plus CPU/GPU limits and audit trails so non‑developer users (HR, marketing) can safely run agents.
AI-written summary · about
chapters (7)
- 0:00Closing panel intro
- 1:58Today’s insights
- 6:33AI sovereignty and withdrawal
- 12:19AI agent costs vs governance
- 18:20Platform engineering beyond devs
- 23:19Building future-proof platforms
- 27:12Closing remarks
related talks
28mClosing panel: The future of platform engineering and what to build next
38mPlatform Engineering in the Age of AI Teammates
32mThe Past, the Present, and the Future of Platform Engineering
31mThe Evolution to Platform Engineering 2.0
01h 02mPlatform Engineering in the age of Generative AI
49mBeyond Engineering: The Future of Platforms
21mPlatforms for Humans and Machines: Engineering for the Age of Agents
34mPlatforms: Defying the laws of IT physics