Notes from the bench

What I'm trying, what broke, what shipped.

A running notebook from the studio — build notes, rough essays, and working ideas across voice, agents, screen-aware tools, XR, and the plumbing in between. Written as the work happens.

Dark interface map showing Scroll as Investigation #1 with voice, screen context, velocity, and attention alignment nodes.
ActiveInvestigation plan8 min read

Scroll as Investigation #1

A working investigation into ambient and voice-based agent interaction: screen context, scroll velocity, narration, attention alignment, and real-time questions.

Read the note →

Dark evaluation matrix for agent interfaces with cognitive load, trust, interruption, and recovery axes.
DraftEssay9 min read

The Case for Human-Factors-Grade Agent Interfaces

Why future agent interfaces need more than pretty components: legibility, cognitive load, trust calibration, and reproducible evaluation.

Read the note →

Decision tree for agent interface selection with branches for text, voice, form, dashboard, timeline, and XR workspace.
BacklogField note7 min read

When Should an Agent Generate an Interface?

A decision tree for moving from chat answers to generated forms, timelines, maps, workspaces, and spatial control panels.

Read the note →