Skip to content
John HiemstraCreative
John HiemstraCreative

Product Engineering

Solutions made to measure.

I'm John Hiemstra, hybrid problem solver & product engineer/owner, building an agentic practice that ships apps, sites and systems solving real business challenges.

See the work
… appreciate his ability to make the connection between the business and the systems… to understand, design, and implement applications that solve actual business problems.
— Neil Olson, Engineering Leader

How I work

John Hiemstra - Product Engineer

I learned product engineering on enterprise teams: designing product workflows, building component libraries shared across applications, shipping web apps intended for 500,000 concurrent users and leading a $40M-a-year legacy-to-cloud move with no customer disruption.

Those teams shipped products users could trust because we had a process we could trust, and we fixed it every time we got something wrong.

Now I run that process with a team of AI agents. Each role works to shared policies, a person reviews the work at every gate, and every correction goes back into the policies, so the agents improve with every rep.

What we're building

Tailor is an application generator we built this summer to replace WordPress and the WP sites we've maintained. Designed & owned by me; built, tested and shipped by agents. No plugins to patch, no update cycle to babysit, no PHP. Even heavy pages score 92–100 on Lighthouse, and answer engines can read every page, blog post and product.

Fast on a phone.

Google's Lighthouse scores how fast a page loads on a phone, from 0 to 100. We ran it the same afternoon on CNN, ESPN and the BBC, on Rockstar Blends before it moved, and on two Tailor sites. Both Tailor sites score in Google's good band: 96 for the Rockstar Blends shop and 100 for this site.

What else makes a Tailor site →

Lighthouse mobile performance scoreLighthouse mobile, three runs each, November 2, 2026 (BigCommerce: June 2026). CNN: score 43, 10.8 seconds; ESPN: score 37, 7.4 seconds; BBC: score 53, 4.3 seconds; Rockstar Blends: score 70; Rockstar Blends on Tailor: score 96, 2.6 seconds; johnhiemstra.com on Tailor: score 100, 1.4 seconds. Google rates a score of 90 to 100 good, 50 to 89 needs improvement and under 50 poor; and the largest content on screen in 2.5 seconds or less good, up to 4 seconds needs improvement, past 4 poor.Lighthouse performance score on a phonePoorNeeds workGoodCNN43ESPN37BBC53Rockstar Blends (subscription, 6/16/26)70Rockstar Blends on Tailor96johnhiemstra.com on Tailor1000204050608090100Largest Contentful Paint, secondsGoodNeeds workPoorCNN10.8 sESPN7.4 sBBC4.3 sRockstar Blends on Tailor2.6 sjohnhiemstra.com on Tailor1.4 s0 s2.5 s4 s6 s8 s10 s12 sLighthouse performance score on a phone, against big media sitesLighthouse mobile, three runs each, November 2, 2026 (BigCommerce: June 2026). CNN: score 43, 10.8 seconds; ESPN: score 37, 7.4 seconds; BBC: score 53, 4.3 seconds; Rockstar Blends: score 70; Rockstar Blends on Tailor: score 96, 2.6 seconds; johnhiemstra.com on Tailor: score 100, 1.4 seconds. Google rates a score of 90 to 100 good, 50 to 89 needs improvement and under 50 poor; and the largest content on screen in 2.5 seconds or less good, up to 4 seconds needs improvement, past 4 poor.Lighthouse performance score on a phonePoorNeeds workGoodCNN43ESPN37BBC53Rockstar Blends (BigCommerce, June)70Rockstar Blends on Tailor96johnhiemstra.com on Tailor10005090100Largest Contentful Paint, secondsGoodNeeds workPoorCNN10.8 sESPN7.4 sBBC4.3 sRockstar Blends on Tailor2.6 sjohnhiemstra.com on Tailor1.4 s0 s2.5 s4 s8 s12 s

Managing the LCP

Largest Contentful Paint is how long you wait for the biggest element on the page to load, often an image or other media asset. It's generally the moment the page looks & feels 'ready' for you. Google rates 2.5 seconds or less as 'Good'. On a typical 4G connected phone, this site shows is generally 'ready' around 1.5 seconds; CNN around 10.

And saved money doing it →

Largest Contentful Paint on a phone, in seconds, against big media sitesLighthouse mobile, three runs each, November 2, 2026 (BigCommerce: June 2026). CNN: score 43, 10.8 seconds; ESPN: score 37, 7.4 seconds; BBC: score 53, 4.3 seconds; Rockstar Blends: score 70; Rockstar Blends on Tailor: score 96, 2.6 seconds; johnhiemstra.com on Tailor: score 100, 1.4 seconds. Google rates a score of 90 to 100 good, 50 to 89 needs improvement and under 50 poor; and the largest content on screen in 2.5 seconds or less good, up to 4 seconds needs improvement, past 4 poor.Lighthouse performance score on a phonePoorNeeds workGoodCNN43ESPN37BBC53Rockstar Blends (Subscription, 6/16/26)70Rockstar Blends on Tailor96johnhiemstra.com on Tailor1000204050608090100LCP in secondsGoodNeeds workPoorCNN10.8 sESPN7.4 sBBC4.3 sRockstar Blends on Tailor2.6 sjohnhiemstra.com on Tailor1.4 s0 s2.5 s4 s6 s8 s10 s12 sLargest Contentful Paint on a phone, in seconds, against big media sitesLighthouse mobile, three runs each, November 2, 2026 (BigCommerce: June 2026). CNN: score 43, 10.8 seconds; ESPN: score 37, 7.4 seconds; BBC: score 53, 4.3 seconds; Rockstar Blends: score 70; Rockstar Blends on Tailor: score 96, 2.6 seconds; johnhiemstra.com on Tailor: score 100, 1.4 seconds. Google rates a score of 90 to 100 good, 50 to 89 needs improvement and under 50 poor; and the largest content on screen in 2.5 seconds or less good, up to 4 seconds needs improvement, past 4 poor.Lighthouse performance score on a phonePoorNeeds workGoodCNN43ESPN37BBC53Rockstar Blends (BigCommerce, June)70Rockstar Blends on Tailor96johnhiemstra.com on Tailor10005090100Largest Contentful Paint, secondsGoodNeeds workPoorCNN10.8 sESPN7.4 sBBC4.3 sRockstar Blends on Tailor2.6 sjohnhiemstra.com on Tailor1.4 s0 s2.5 s4 s8 s12 s

Are the agents looking you up?

When someone asks ChatGPT, Claude or Perplexity about what you sell, the answer engine fetches pages to answer. For 30 days, AI answer engines fetched Rockstar Blends pages every day, never fewer than 24 times. Tailor gives them the same page a person gets, in full, with an index written for agents, and its Workbench shows you who came looking.

The three readers →

AI answer fetches a day, October 4 to November 2, 2026Line over 30 days on a scale from zero to 100. Every day AI answer engines fetched pages to answer someone: never fewer than 24, at most 82, 1,387 in all, and 65 on November 2.AI fetching a page to answer a question, a day050100Oct 4Nov 265AI answer fetches a day, October 4 to November 2, 2026Line over 30 days on a scale from zero to 100. Every day AI answer engines fetched pages to answer someone: never fewer than 24, at most 82, 1,387 in all, and 65 on November 2.AI answer fetches a day050100Oct 4Nov 265
… able to communicate very new ideas and difficult concepts to our staff — energetic, fun to work with, and optimistic about the possibilities for improvement.
— Laura Baker, Administrative Leader

Services

Let's talk.

Book a time
  • Bloom Integrated Health
  • ESPN - John Hiemstra Creative
  • PwC - John Hiemstra Creative
  • Rockstar Blends - John HIemstra Creative
  • City of San Diego - John Hiemstra Creative
  • Havas - John Hiemstra Creative
  • BBDO - John Hiemstra
  • Sapient Nitro - John Hiemstra Creative
  • Edelman - John Hiemstra Creative
  • heather reilly hiemstra, metalsmith - John Hiemstra Creative
  • Las Vegas CVB - John Hiemstra Creative
  • LocalVox - John Hiemstra Creative
  • History Channel - John Hiemstra Creative
  • Digital Pulp
  • The Picket Fence Pittsburgh - John Hiemstra Creative
  • San Francisco International Airport - John Hiemstra Creative
  • Conteneo - John Hiemstra Creative
  • Scaled Agile - John Hiemstra Creative
  • Code & Theory
  • CORO Pittsburgh - John Hiemstra Creative
  • DrFirst - John Hiemstra Creative
  • IATSE Local 816 - John Hiemstra Creative