Field Notes, page 9 — Design Outcomes

DESIGN SYSTEMS COLLECTIVE

LDLR DESIGN FIELD NOTES

Agents don’t read docs

Agents read structure and naming, not docs or goodwill, and the informal contract becomes a hard boundary.

MAY 11, 2026

Systems & primitives Craft & detail

MICROSOFT DESIGN

LDLR DESIGN FIELD NOTES

The system is the handoff

Microsoft folded Copilot into four surfaces tied by one ‘Throw and Catch’ pattern: the pattern is the system.

BY JON FRIEDMAN / MAY 12, 2026

A simplified system

Intelligence without continuity can feel like an interruption instead of a thought partner.

Why it matters

— Four canonical Copilot surfaces, not scattered features per app.

— 'Throw and Catch' governs handoff between the four surfaces.

— Context persists across apps. The conversation does not restart.

— On-canvas narrows from exploration to document focus on selection.

Read on Microsoft Design →

Systems & primitives Prototyping & exploration Craft & detail

GITHUB (ENGINEERING & DESIGN)

LDLR DESIGN FIELD NOTES

Instant is the spec

GitHub Issues reframed performance around distribution shape, then paid 4.7% server/cache divergence to feel instant.

BY ALEXANDER LELIDIS / MAY 14, 2026

From latency to instant: Modernizing GitHub Issues navigation performance

We are willing to serve data that may be slightly stale if that allows the navigation to complete near instantaneous

Why it matters

— Reframed the goal: shape the distribution, not chase p99.

— Five layers: in-memory cache, IndexedDB, preheat, service worker, code-split.

— Accepted 4.7% server/cache divergence as the operating envelope.

— Cache hit went from 33% to 96%. P10 dropped to 70ms.

Read on GitHub (Engineering & Design) →

Craft & detail Systems & primitives Critique & decision-making

DESIGN SYSTEMS COLLECTIVE

LDLR DESIGN FIELD NOTES

AI didn't add to the spec

AI-readiness is the old spec finally enforced: tokens, layout intent, and governance now have to read to a non-human.

APR 27, 2026

Legible by Design: What AI Readiness Actually Demands of a Design System

The best systems are the ones that remain coherent even when the people who built them have moved on

Why it matters

— AI doesn't need a new system. It exposes the broken one.

— Five load-bearing properties: tokens, layout intent, a11y, governance, handoff.

— Legibility to machines reveals what was always opaque between humans.

— Stewardship beats generation. The system has to outlive its authors.

Read on Design Systems Collective →

Systems & primitives Critique & decision-making Cross-functional & process

WEBFLOW

LDLR DESIGN FIELD NOTES

Where AI debt lands

AI code defers cost instead of cutting it: seven debt categories land far from where the velocity gain was booked.

BY ANKIT AGRAWAL / MAY 12, 2026

The cleanup cost of ungoverned AI-generated code

It costs a contributor five minutes to generate a low-quality AI pull request, and hours for the maintainer to verify and reject it.

Why it matters

— Seven debt categories, segmented by who actually pays the bill.

— Time-to-exploit window collapsed from months to days.

— Five minutes to generate a PR. Hours to reject one.

— Triage PRs by risk class. Output is the wrong metric.

Read on Webflow →

Critique & decision-making Cross-functional & process Systems & primitives

MICROSOFT DESIGN

LDLR DESIGN FIELD NOTES

The missing seat

Frontier AI teams optimize what’s measurable, so Microsoft’s researchers want a seat where the metrics get chosen.

BY MARTIN LINDE DANTY, RABEEZA A., ANNE KLÆRKE-OLESEN, MADHUMAY SINHA, MICHAEL O'SULLIVAN / APR 28, 2026

The missing seat at the frontier team table

When evaluation is driven primarily by available data, teams risk optimizing what is measurable rather than what actually matters.

Why it matters

— User research belongs in evaluation, not adjacent to it.

— Session-length looked strong. Task-completion cut repeat queries by a third.

— Sales-order workflow assumed isolation. Research surfaced cascading approvals.

— Concept tests caught a workflow misread that would have cost weeks.

Read on Microsoft Design →

Research & validation Cross-functional & process Critique & decision-making

FIGMA

LDLR DESIGN FIELD NOTES

The canvas reads code

McMillin reframes canvas, code, and PRD as one system: an agent translates, and judgment moves through both.

BY BRETT MCMILLIN / APR 30, 2026

Workflow lab: Expanding the canvas with Figma MCP

The artifact isn't a bug report. It's a shared reference for a design conversation.

Why it matters

— Agents surfaced ten product states the original canvas missed.

— Visual diff replaces the bug ticket as a design artifact.

— On canvas exploration stays cheap. In code every direction costs.

— Source of truth becomes connected artifacts, not one canonical file.

Read on Figma →

Prototyping & exploration Cross-functional & process Systems & primitives

LINEAR

LDLR DESIGN FIELD NOTES

Hiring, written down

Writing down how you hire forces honesty about what you value. Linear did the work and published it.

BY DOUG PARKER / APR 28, 2026

Hiring & team Cross-functional & process Critique & decision-making

MICROSOFT WORKLAB

LDLR DESIGN FIELD NOTES

Workers ready, orgs not

Microsoft’s 2026 Work Trend Index, 20,000 workers across ten countries: only 26 percent see leadership aligned on AI.

BY MICROSOFT WORK TREND INDEX TEAM / MAY 5, 2026

Agents, human agency, and the opportunity for every organization

Workers are ready. Their organizations aren't.

Why it matters

— One in five workers sit in the Frontier zone.

— One in ten are ready individuals blocked by the organization.

— Only twenty-six percent say leadership is consistently aligned on AI.

— Sixty-five percent fear falling behind. Forty-five percent stay in old work.

Read on Microsoft WorkLab →

Cross-functional & process Research & validation Storytelling & narrative

LINEAR

LDLR DESIGN FIELD NOTES

Kept in their lane

A Linear engineer scopes agent tasks tight enough that they can’t wander, the same discipline as scoping for a junior.

How we use Linear Agent at Linear

Matthijs gets the best results from coding agents by breaking the work into small, targeted steps.

Why it matters

— Scoping for an agent is the same discipline as scoping for a junior.

— Three-team workflow shows where agents add real leverage.

— PM-ships-the-fix works because the change was small, not the agent.

— Marketing-shaped piece. The underlying tactics are still real.

Read on Linear →

Cross-functional & process Critique & decision-making

MAGGIE APPLETON

LDLR DESIGN FIELD NOTES

The design bottleneck

Appleton reads Yegge’s chaotic agent orchestrator as design fiction: the coding bottleneck fell, the design one stayed.

JAN 23, 2026

Gas Town’s Agent Patterns, Design Bottlenecks, and Vibecoding at Scale

Design becomes the limiting factor: imagining what you want and figuring out the gnarly details to make it real.

Why it matters

— When agents write faster than you design, vibe-designing is a footgun.

— Mayor/Polecat hierarchy sketches a real coordination pattern for agents.

— Code distance shifts with domain, stakes, and team size.

— Infrastructure matters more than models. Validation loops, not smarter agents.

Read on their blog →

Critique & decision-making Cross-functional & process Prototyping & exploration

MAGGIE APPLETON

LDLR DESIGN FIELD NOTES

The PR can’t plan

Agents collapsed the implementation window, and the alignment checkpoints that lived inside it went with it.

APR 13, 2026

One Developer, Two Dozen Agents, Zero Alignment

Code is so cheap we don't properly stop to think before prompting it. We lose more alignment points.

Why it matters

— Planning and building are no longer separate phases. The tools still are.

— The PR now carries upstream checkpoints it was never designed for.

— Most alignment context lives in people's heads. Agents can't reach it.

— Ace is a research prototype. The diagnosis is sharper than the fix.

Read on their blog →

Cross-functional & process Critique & decision-making Prototyping & exploration