Field Notes — Design Outcomes

SHOPIFY ENGINEERING

LDLR DESIGN FIELD NOTES

Reliable convergence

Naqvi gates LLM migration work checkpoint by checkpoint: prove behavior, review the pixels, then a human signs.

BY TALHA NAQVI / SEP 21, 2026

Helix: The internal tool powering our Shopify app’s native migration

We stopped optimizing for a perfect first attempt and started working towards reliable convergence.

Why it matters

— A 300-screen React Native app, rebuilt native by LLMs in small checkpoints

— Four gates in order: CLI-proven behavior, visual review, adversarial code review, engineer approval

— Gemini plays perfectionist design reviewer, comparing screenshots against the React Native reference

— Engineer feedback feeds a memory that grows autonomy as the migration advances

Read on Shopify Engineering →

Cross-functional & process Critique & decision-making

MICROSOFT DESIGN

LDLR DESIGN FIELD NOTES

The empty room

Jiang brings Sunyata to Copilot: automation as emptying, removing mechanical labor to make room, not adding more.

BY QIAN JIANG / SEP 22, 2026

Create Ways of Being

Emptiness is the space in which reality can reveal itself.

Why it matters

— An AI designer on Copilot who has shown work at the Louvre and Tate Modern

— From Zhuangzi and Kant to Alipay sweet potatoes: hyper-digitization exhausts what it connects

— Sunyata as design principle: automation removes mechanical labor instead of adding features

— Her Garden of Synthetic Eden reimagines Bosch through AI’s birth and failure modes

Read on Microsoft Design →

Storytelling & narrative Craft & detail

PROOF OF CONCEPT

LDLR DESIGN FIELD NOTES

Be the stakeholder

Hoang’s case for the design GM: own a P&L, run the roster, stop appeasing the stakeholder and be one.

SEP 20, 2026

The Design GM

It’s time to go from appeasing stakeholders to being the stakeholder.

Why it matters

— The GM builds the roster and owns the outcome; the coach wins the games

— Three areas: business outcomes, operations that earn their calendar, a cross-disciplinary roster

— At Replit he owned marketing and the self-serve funnel, signup through renewal

— Few GMs come from design; Bruce Bell at Square is the known exception

Read on Proof of Concept →

Hiring & team Cross-functional & process

NIELSEN NORMAN GROUP

LDLR DESIGN FIELD NOTES

Curation beats prompting

NN/g’s power-user study lands one shift: context curation, not prompt craft, is where the leverage moved.

BY TANNER KOHLER / SEP 18, 2026

The 3 Roles of Context for AI Agents

Curating context now matters more than writing the perfect prompt.

Why it matters

— Every power user studied had built a context library before agents got good

— Three roles: global rarely changes, local tracks the task, ambient streams in raw

— Handling follows role: curate global, let the agent write local, let it sift ambient

— A deliberately lazy prompt worked fine because the MD file did the heavy lifting

Read on Nielsen Norman Group →

Research & validation Systems & primitives

THE MERHOLZ AGENDA

LDLR DESIGN FIELD NOTES

Directors own the how

Merholz maps why AI lands hardest on UX Directors: the mandate is the How, and the How is theirs alone.

SEP 21, 2026

The Ongoing Plight of the UX Director

Executives own the Why, Directors the How, and AI adoption is the How.

Why it matters

— AI adoption is tooling and process, the How, and the How belongs to Directors

— Mandates arrive with no guidance during the worst technology rollout in history

— The ratio math: a Director over 20 sits peered against a VP over 160

— The 2021 hyperscaling cohort is stuck: orgs flatten and VP roles vanish

Read on The Merholz Agenda →

Hiring & team Cross-functional & process

GITHUB

LDLR DESIGN FIELD NOTES

Two geometries

Gimeno opens a million-line PR without dropping a frame: one geometry for code, a second for everything human.

BY ALBERTO GIMENO / SEP 23, 2026

Rendering huge pull requests in the GitHub Copilot app

Any ‘is the user interacting?’ check has to be one your own side effects can’t satisfy.

Why it matters

— A 2,200-file, million-line PR opens scrollable, with 400 inline review comments

— Two geometries: deterministic prefix sums for code, measured dynamic blocks for comments

— They rejected per-block ResizeObservers for one idle-gated measurement pass near the viewport

— An autopilot drove the real app unattended, reading health signals from disk logs

Read on GitHub →

Craft & detail Prototyping & exploration

BUILDING NUBANK

LDLR DESIGN FIELD NOTES

Written to be heard

A low-vision Nubanker wrote BLIND-CLAUDE.md: every response must survive being read aloud, top to bottom, in order.

BY NUBANK EDITORIAL / SEP 16, 2026

Designing Claude for ears, not eyes: An accessibility skill for blind and low-vision users

Every response must make sense read top-to-bottom, out loud, in order.

Why it matters

— BLIND-CLAUDE.md treats auditory comprehension as an engineering constraint, not a UI afterthought

— Tables become prose: a screen reader narrates cells as disconnected values with no structure

— Spatial references like 'see above' become restatements; every section carries its own context

— Long build logs collapse to the two or three lines that matter

— One person's validated workflow, honestly flagged: no broader user feedback collected yet

Read on Building Nubank →

Craft & detail Systems & primitives

DESIGN SYSTEMS COLLECTIVE

LDLR DESIGN FIELD NOTES

Can, should, may

Ahmadi's test for every agent action: execution, judgment, governance. Readiness is legibility, not tooling.

SEP 16, 2026

A Design System Isn't AI-Ready Until Its Decisions Are Legible

Can the agent do it? Should it do it this way? May it decide this?

Why it matters

— AI readiness is a governance problem before it is ever a tooling problem

— An agent invents the missing button variant and commits it to the shared library

— Decision legibility: a non-participant sees what was decided, why, and who owns it

— Source of truth says what exists; source of authority says why it binds

— Agent experience design: UX for the repository the agent has to navigate

Read on Design Systems Collective →

Systems & primitives Critique & decision-making

DESIGN SYSTEMS COLLECTIVE

LDLR DESIGN FIELD NOTES

The bill under the demo

45 seconds to generate the screen, three hours to make it honest. Singh's metric: time to trusted output.

SEP 18, 2026

The 'Fix-it Tax' of Generative UI

It is a fast proposal with a bill folded underneath.

Why it matters

— Green rectangles for switches, raw hex for surface.base, focus states stripped for cleanliness

— Generated UI delivers the appearance of decisions without the decision trail

— Cleanup is archaeology: find the assumption, trace the rule, replace it without breaking structure

— Her fix: a pattern contract written before generation, tokens and boundaries included

— One innocent token had failed contrast testing six months earlier; no model knows that

Read on Design Systems Collective →

Systems & primitives Prototyping & exploration

UX COLLECTIVE

LDLR DESIGN FIELD NOTES

The second collision

Neeman reads Nader into AI: the model's error is the first crash, the interface's handling of it is yours.

SEP 18, 2026

Altman and Amodei's 'unsafe at any prompt' moment for AI

Capability has a leaderboard. Harm has a mailing list.

Why it matters

— 'Prompt better' is Detroit's nut behind the wheel defense, reframed for AI

— Stanford's 2026 Index: hallucination rates from 22 to 94 percent across 26 models

— The metric to steal: wrong-answer survival rate, measured with seeded failures

— Pad the interior: citations, visible uncertainty, cheap undo before Article 86 forces it

Read on UX Collective →

Critique & decision-making Research & validation

NIELSEN NORMAN GROUP

LDLR DESIGN FIELD NOTES

The byline still binds

NN/g's editor opens the workflow: AI drafts, critiques, compresses. A human expert still signs every claim.

BY RALUCA BUDIU / SEP 11, 2026

AI Can Help Write an Article, but It Can't Stand Behind It

AI may contribute words and even ideas, but humans remain responsible for editorial judgment.

Why it matters

— The EU AI Act exempts content under human review; that clause is the hinge

— AI rewrites for concision, hits the 160-character summary, adapts reports into articles

— ChatGPT caught a contradiction in an earlier draft of this very article

— The kicker: NN/g's AI content policy was itself AI-drafted, then human-approved

Read on Nielsen Norman Group →

Cross-functional & process Storytelling & narrative

UX COLLECTIVE

LDLR DESIGN FIELD NOTES

A hundred sharpenings

Yeo's 100th article, with a confession: editors caught his AI drafts. The friction of repetition is the point.

SEP 17, 2026

Rethinking writing in the world of design and AI

If you cannot write well, others will do your thinking for you.

Why it matters

— He tried AI-generated drafts; editors who knew his voice caught the slop

— Bierut's 100-Day logic: the friction of repetition, not the output, grows the writer

— The practice is specific: Google Keep drafts on his phone while kids fall asleep

— Seven lessons for sustaining a practice, from motivation and voice through forgiveness and tribe

Read on UX Collective →

Storytelling & narrative Craft & detail