Coco desktop agent learns when to help via a continual personalization loop
EchoShao8899 · x · 2026-11-16
EchoShao describes Coco, a proactive desktop agent whose core challenge is the same as a human assistant's: knowing when a suggestion helps versus when it interrupts—citing this as the bottleneck for most personal agent products.
Deep Personalization works as a continual learning loop (architecture shown in a diagram):
- Observe: an on-device Observer agent follows workflows across apps, tabs, and activities, keeping compact structured contextual memory (observation, userintent, rationale, needsupport).
- Learn: help requests, recurring patterns, and Accept/Ignore/Modify feedback on suggestions are treated as training signal.
- Adapt: the Observer is updated via evolutionary prompt optimization or LoRA fine-tuning, with evaluation against an OSWorld-Persona benchmark for intervention quality and personalization.
Through everyday interaction it learns when each user wants assistance.
Related event: Coco Desktop Agent Learns When to Proactively Help Users(5 posts)→
More from coding & agent
- OSWorld-Persona: a 3,259-label benchmark for evaluating when agents should proactively help — EchoShao8899 · 2026-11-16
- Let Claude write skills for your local Flash model — it works surprisingly well — gnukeith · 2026-11-16
- Google: Hackers Now Run Autonomous Multi-Agent Frameworks to Weaponize Patches — AccBalanced · 2026-11-16
- Google Chrome hiring exec to build the agentic browser era — laparisa · 2026-11-16
- Cursor 'kinda dead' 3 months post-acquisition, says HF's Niels Rogge — NielsRogge · 2026-11-16
- DJI Mic Mini recorder app built with Codex in 3-4 prompts, shared by OpenAI Devs — OpenAIDevs · 2026-11-16