Skip to content

Instantly share code, notes, and snippets.

View StijnZanders's full-sized avatar

Stijn Zanders StijnZanders

View GitHub Profile
@StijnZanders
StijnZanders / ai-triage-gist.md
Created April 26, 2026 12:46
AI Triage UI - a spec for building keyboard-driven human-in-the-loop decision interfaces

AI Triage UI

A spec for building keyboard-driven human-in-the-loop decision UIs. Give this to any AI coding tool and it should produce a working prototype.

The idea: AI prepares decisions, humans make the call. Every override teaches the system something.

What you're building

A card stack. Each card is one case that needs a human decision. The AI has already done the investigation: scored the case, surfaced the key signals, written a draft action (email, routing decision, approval, whatever fits your domain). The human reviews and either accepts or overrides.

@StijnZanders
StijnZanders / ai-triage-command-center.md
Created April 26, 2026 07:13
AI Triage Command Center - a UI pattern for keyboard-driven, human-in-the-loop decision-making with multi-agent consensus

AI Triage Command Center

A UI pattern for human-in-the-loop decision-making at speed. Multiple AI agents analyze each case from different angles, present a consensus with a recommendation, and a human operator accepts or overrides using keyboard shortcuts. Think "Tinder for ops decisions" - card stack, swipe to decide, arrow keys to fly through a queue.

The Core Idea

Most AI-assisted workflows fall into one of two traps: either the AI decides everything (dangerous) or the human reviews everything from scratch (slow). This pattern hits the middle: AI does the analysis, human makes the call.

The key insight is that a well-designed triage UI can compress a 15-minute investigation into a 5-second decision. The AI pre-digests the evidence. The human just needs to agree or disagree.