The End of Apps
summary
Argues that apps fragment tasks and predicts an agentic OS that runs 95% of work and nudges humans only for decisions. Describes building Benji, Android Tasker automations, voice capture, Claude Code / OpenClaw / Clawbot experiments, and failures of one‑to‑one chat. Shows concrete workflows for email, home, finance, and booking, plus a surprising number of tiny automations that worked. Proposes temporary interfaces and portable skills with Executor and Skillbox as the next layer.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:14A lifetime of productivity systems
- 2:00Android Tasker and voice capture
- 3:00From Notion life OS to Benji
- 5:00When ChatGPT plugins changed the interface
- 6:00An early AI voice assistant
- 7:00Claude Code beyond coding
- 8:50Discovering Clawbot
- 10:00Personal-agent use cases
- 11:00Calls, email security, and booking real-world work
- 12:00Home automation and personal data
- 14:00Native app automation
- 15:00Podcasts, wearables, and Android control
- 16:45When the agent honeymoon ended
- 17:22Why one-to-one agent chat fails
- 18:00Too many agents and too many chats
- 19:00Cloud agents versus self-hosting
- 20:00Temporary apps inside conversations
- 21:00Portable tools with Executor and Skillbox
- 21:50Toward an agentic operating system
- 22:20Reversing the role of people and AI
- 23:00Which apps survive
related talks
20mThe End of Apps
24mThe 'A' Revolution: Apps, APIs & AI - a Singularity
35mVibe Code Your Personal Operating System | AI Productivity with Bethany Crystal
23mPersonal OS: Build Your Army of 100 AI Interns
41mHow AI Will Change Mobile Development
18m"Software Fundamentals Matter More Than Ever"
44mGo Beyond the Prompt: Using Context to Build Reliable AI-Assisted Android Workflows
46mAI Agents Are Becoming Operating Systems: Lessons From OpenClaw