# Blueprints are out, Wind tunnels are in

- Author: @gurupanguji
- Date: 2025-08-18
- URL: https://gurupanguji.com/blog/2025/08/18/blueprints-are-out-wind-tunnels-are-in/
- Categories: essay, ai
- Tags: ai, ai-in-product-managemement, design, eng, engineering, llm-prototyping, llm, okrs, pm, pm-ux-engineering, prd, product, product-management, product-lifecycle, prototyping, startups, ux, uxr, velocity, visual-design

---


<!-- wp:heading -->
<h2 class="wp-block-heading">LLMs collapse the product loop</h2>
<!-- /wp:heading -->

<!-- wp:paragraph -->
<p>Two weeks, two side projects, two very different partners: one deep in code, the other steeped in business ops. Building with both made something obvious: large language models aren’t just engineering accelerants. <em>They collapse the entire product loop.</em></p>
<!-- /wp:paragraph -->

<!-- wp:paragraph -->
<p>For the past decade, most teams have organized around a trinity - PM, UX, Eng. Because engineering was scarce and expensive, we optimized for engineering leverage. The culture drifted toward documents and handoffs: PRDs, mocks, design docs, reviews. “Handoff” became the milestone; building became the after-party. As a result, learning from customers slid months to the right.</p>
<!-- /wp:paragraph -->

<!-- wp:paragraph -->
<p>Somewhere in there, “throwaway work” became a slur. Prototyping without a path to production was treated as waste. The best engineers quietly ignored that taboo, hacked something together, and came back with actual signal. Everyone else waited for the next review.</p>
<!-- /wp:paragraph -->

<!-- wp:paragraph -->
<p>LLMs change the math. They parallelize the trinity and radically reduce the cost of being wrong.</p>
<!-- /wp:paragraph -->

<!-- wp:list -->
<ul class="wp-block-list"><!-- wp:list-item -->
<li><strong>PM</strong> can draft a crisp one-pager, expand risks and edge cases, and generate three experiment plans in an afternoon.</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li><strong>UX</strong> can turn that one-pager into clickable flows and alternative microcopy the same day.</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li><strong>Engineering</strong> can scaffold a fake backend, wire in guardrails, and instrument a demo by day two.</li>
<!-- /wp:list-item --></ul>
<!-- /wp:list -->

<!-- wp:paragraph -->
<p>The point isn’t that AI writes production code; it shrinks <strong>time-to-signal.</strong> Treat prototypes as wind-tunnel models: not for flying, for learning how air hits your wing.</p>
<!-- /wp:paragraph -->

<!-- wp:heading -->
<h2 class="wp-block-heading">A faster loop, by role</h2>
<!-- /wp:heading -->

<!-- wp:list -->
<ul class="wp-block-list"><!-- wp:list-item -->
<li><strong>PM</strong>: Use LLMs to create the first PRD draft, enumerate unknowns, and produce testable hypotheses. Generate interview guides and summary memos so insights move.</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li><strong>UX</strong>: Translate hypotheses into mid-fi flows, produce microcopy variants, and run task simulations. Build a click/tap-through that captures the core choice you need users to make.</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li><strong>Eng</strong>: Ask the model for scaffolding, contract tests, and a stubbed data layer. Auto-generate a design-doc outline from code comments as you go. Instrument from day zero.</li>
<!-- /wp:list-item --></ul>
<!-- /wp:list -->

<!-- wp:paragraph -->
<p>Run a 72-hour <strong>Learning Sprint:</strong> build just enough to put in front of five users. Kill fast or iterate. “Throwaway” code is paid research.</p>
<!-- /wp:paragraph -->

<!-- wp:heading -->
<h2 class="wp-block-heading">Change the scoreboard</h2>
<!-- /wp:heading -->

<!-- wp:paragraph -->
<p>If we continue to (over) optimize for engineering utilization, we will keep writing documents. If we optimize for learning, we will ship smaller bets - faster. Here are some metrics that might help you think about it better: </p>
<!-- /wp:paragraph -->

<!-- wp:list -->
<ul class="wp-block-list"><!-- wp:list-item -->
<li>cycle time for a validation signal</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>learning velocity (meaningful insights / week)</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>percent of bets killed quickly (and cycle time to bet killing)</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>prototype half-life (time until learnings hit prod)</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>alignment drift (features without a link to a top level outcome)</li>
<!-- /wp:list-item --></ul>
<!-- /wp:list -->

<!-- wp:paragraph -->
<p>Make OKRs smaller and more numerous. Measure the loop, not the launch. Celebrate the team that invalidates shiny idea in 3 days. </p>
<!-- /wp:paragraph -->

<!-- wp:heading -->
<h2 class="wp-block-heading">Regulated shouldn't mean slow</h2>
<!-- /wp:heading -->

<!-- wp:paragraph -->
<p>Finance and health have longer cycles, but the pattern should hold. </p>
<!-- /wp:paragraph -->

<!-- wp:list -->
<ul class="wp-block-list"><!-- wp:list-item -->
<li>Try compliance sandboxes with anonymized / synthetic data; log prompts and outputs</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>Learn from the crypto gang: let's enable policy-as-code gatekeeping for prototypes</li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>Generate pre-read packs for the legal and compliance gang to bring them along </li>
<!-- /wp:list-item -->

<!-- wp:list-item -->
<li>Get compliance to be part of witness tests to get them to break the prototype </li>
<!-- /wp:list-item --></ul>
<!-- /wp:list -->

<!-- wp:paragraph -->
<p>For the long term, we have to bring the regulators along also and we need them to move from blockers -> collaborators. </p>
<!-- /wp:paragraph -->

<!-- wp:heading -->
<h2 class="wp-block-heading">Make product fun again </h2>
<!-- /wp:heading -->

<!-- wp:paragraph -->
<p>The joy of product is the rapid loop: talk to users, build a little, learn a lot. LLMs give that loop back. Use them to parallelize the trinity, to lower the cost of being wrong, and to move learning left. Optimize for time-to-signal, and the rest of the process starts behaving.</p>
<!-- /wp:paragraph -->

<!-- wp:paragraph -->
<p>Ship the loop, not the doc.</p>
<!-- /wp:paragraph -->