Keynote: Otto
summary
Otto is an experimental AI workflow framework built to handle help‑desk tickets by running a task execution loop with four components: trigger, context, instructions, and tools (search docs, read articles, send replies). Tested on about 3500 tickets, it auto‑drops responses for ~70% and costs roughly $700 to run, producing a saved ≈24 work hours/week; user feedback is ~40% positive. It adds shadow replies, automated evaluations comparing AI vs human on helpfulness and tone, and exports tooling for Frappe/ERPNext integration.
AI-written summary · about
chapters (7)
- 0:00Introduction to Otto
- 1:58Core concept behind Otto
- 3:44Handling first responses
- 7:18Evaluating ticket responses
- 9:28Outcomes and learnings
- 11:34Improving documentation quality
- 12:36Otto ecosystem and future
related talks
54mKeynote: AI-Powered App Development
49mKeynote: Operations and AI with Jen Krieger
16mDon't just talk to AI, do more with AI: how to improve productivity with AI agents
01h 00mKeynote Panel AI from Investment perspective
15mHands-on Practice
26mFour Steps to Embed AI Agents into Workflow
24mKeynote - The AI Playbook For Companies
47mAI-Powered Software Development From the Trenches