{"id":4700,"date":"2026-04-21T07:39:39","date_gmt":"2026-04-21T05:39:39","guid":{"rendered":"https:\/\/agentaya.com\/?p=4700"},"modified":"2026-04-21T07:39:52","modified_gmt":"2026-04-21T05:39:52","slug":"zero-coding-skills-100-budget-how-one-entrepreneur-vibe-coded-a-complete-ai-photography-platform-from-scratch","status":"publish","type":"post","link":"https:\/\/agentaya.com\/nl\/zero-coding-skills-100-budget-how-one-entrepreneur-vibe-coded-a-complete-ai-photography-platform-from-scratch\/","title":{"rendered":"Zero coding skills, $100 budget: how one entrepreneur vibe-coded a complete AI photography platform from scratch"},"content":{"rendered":"<style>.kb-row-layout-id4700_39d3bf-dd > .kt-row-column-wrap{align-content:start;}:where(.kb-row-layout-id4700_39d3bf-dd > .kt-row-column-wrap) > .wp-block-kadence-column{justify-content:start;}.kb-row-layout-id4700_39d3bf-dd > .kt-row-column-wrap{column-gap:var(--global-kb-gap-md, 2rem);row-gap:var(--global-kb-gap-md, 2rem);max-width:var( --global-content-width, 1290px );padding-left:var(--global-content-edge-padding);padding-right:var(--global-content-edge-padding);padding-top:var(--global-kb-spacing-sm, 1.5rem);padding-bottom:var(--global-kb-spacing-sm, 1.5rem);grid-template-columns:minmax(0, calc(30% - ((var(--global-kb-gap-md, 2rem) * 1 )\/2)))minmax(0, calc(70% - ((var(--global-kb-gap-md, 2rem) * 1 )\/2)));}.kb-row-layout-id4700_39d3bf-dd > .kt-row-layout-overlay{opacity:0.30;}@media all and (max-width: 1024px){.kb-row-layout-id4700_39d3bf-dd > .kt-row-column-wrap{grid-template-columns:repeat(2, minmax(0, 1fr));}}@media all and (max-width: 767px){.kb-row-layout-id4700_39d3bf-dd > .kt-row-column-wrap{grid-template-columns:minmax(0, 1fr);}}<\/style><div class=\"kb-row-layout-wrap kb-row-layout-id4700_39d3bf-dd alignnone wp-block-kadence-rowlayout\"><div class=\"kt-row-column-wrap kt-has-2-columns kt-row-layout-equal kt-tab-layout-inherit kt-mobile-layout-row kt-row-valign-top kb-theme-content-width\">\n<style>.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{display:flex;}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col,.kadence-column4700_0b66f3-92 > .kt-inside-inner-col:before{border-top-left-radius:0px;border-top-right-radius:0px;border-bottom-right-radius:0px;border-bottom-left-radius:0px;}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{column-gap:var(--global-kb-gap-sm, 1rem);}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col > .aligncenter{width:100%;}.kt-row-column-wrap > .kadence-column4700_0b66f3-92{align-self:center;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92{align-self:auto;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col:before{opacity:0.3;}.kadence-column4700_0b66f3-92{position:relative;}@media all and (max-width: 1024px){.kt-row-column-wrap > .kadence-column4700_0b66f3-92{align-self:center;}}@media all and (max-width: 1024px){.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92{align-self:auto;}}@media all and (max-width: 1024px){.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}}@media all and (max-width: 1024px){.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}}@media all and (max-width: 767px){.kt-row-column-wrap > .kadence-column4700_0b66f3-92{align-self:center;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92{align-self:auto;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}.kadence-column4700_0b66f3-92 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}}<\/style>\n<div class=\"wp-block-kadence-column kadence-column4700_0b66f3-92\"><div class=\"kt-inside-inner-col\">\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"200\" height=\"200\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-3.avif\" alt=\"\" class=\"wp-image-4701\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-3.avif 200w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-3-150x150.avif 150w\" sizes=\"auto, (max-width: 200px) 100vw, 200px\" \/><\/figure>\n<\/div><\/div>\n\n\n<style>.kadence-column4700_11070d-81 > .kt-inside-inner-col{display:flex;}.kadence-column4700_11070d-81 > .kt-inside-inner-col,.kadence-column4700_11070d-81 > .kt-inside-inner-col:before{border-top-left-radius:0px;border-top-right-radius:0px;border-bottom-right-radius:0px;border-bottom-left-radius:0px;}.kadence-column4700_11070d-81 > .kt-inside-inner-col{column-gap:var(--global-kb-gap-sm, 1rem);}.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;align-items:center;}.kadence-column4700_11070d-81 > .kt-inside-inner-col > .kb-image-is-ratio-size{align-self:stretch;}.kadence-column4700_11070d-81 > .kt-inside-inner-col > .wp-block-kadence-advancedgallery{align-self:stretch;}.kadence-column4700_11070d-81 > .kt-inside-inner-col > .aligncenter{width:100%;}.kt-row-column-wrap > .kadence-column4700_11070d-81{align-self:center;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81{align-self:auto;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}.kadence-column4700_11070d-81 > .kt-inside-inner-col:before{opacity:0.3;}.kadence-column4700_11070d-81{position:relative;}@media all and (max-width: 1024px){.kt-row-column-wrap > .kadence-column4700_11070d-81{align-self:center;}}@media all and (max-width: 1024px){.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81{align-self:auto;}}@media all and (max-width: 1024px){.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}}@media all and (max-width: 1024px){.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;align-items:center;}}@media all and (max-width: 767px){.kt-row-column-wrap > .kadence-column4700_11070d-81{align-self:center;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81{align-self:auto;}.kt-inner-column-height-full:not(.kt-has-1-columns) > .wp-block-kadence-column.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;}.kadence-column4700_11070d-81 > .kt-inside-inner-col{flex-direction:column;justify-content:center;align-items:center;}}<\/style>\n<div class=\"wp-block-kadence-column kadence-column4700_11070d-81\"><div class=\"kt-inside-inner-col\">\n<p><em>An AgentAya interview with Gilles Storme, co-founder of <\/em><a href=\"https:\/\/www.mastersof.photography\/en\" rel=\"nofollow noopener\" target=\"_blank\"><em>Masters of Photography<\/em><\/a><em> and creator of <\/em><a href=\"https:\/\/www.pixwise.studio\/\" rel=\"nofollow noopener\" target=\"_blank\"><em>Pixwise Studio<\/em><\/a>.<\/p>\n<\/div><\/div>\n\n<\/div><\/div>\n\n\n<p>Gilles Storme has spent over two decades in senior commercial and strategic roles across media, broadcasting, and gaming. He held leadership positions at the BBC, National Geographic, Discovery, and AOL, and later ran commercial teams at King, the studio behind Candy Crush and one of the most successful mobile gaming companies in history. He is, by any measure, a serious and experienced business executive. What he is not, however, is an engineer. He has no coding background. He has never written a line of JavaScript, React, or TypeScript in his life. Yet in the space of a few weeks, working from his home in England, he built and launched <a href=\"https:\/\/www.pixwise.studio\/\" rel=\"nofollow noopener\" target=\"_blank\">Pixwise Studio<\/a>: a fully functioning AI-powered photography learning platform, complete with user accounts, Google social login, a payment system, AI-generated camera settings, AI-powered photo critiques, a community feature, a learning journal, and an SEO content strategy to go with it.<\/p>\n\n\n\n<p>The total cost of external tools and subscriptions? Roughly $100.<\/p>\n\n\n\n<p>If that number sounds almost absurd, that is because it is. Just a few years ago, the same product would have cost somewhere between $30,000 and $80,000 to build with an agency or external development team, with no guarantee it would work or find an audience. Today, a non-technical motivated founder with the right tools and a structured approach can go from idea to live product in days or weeks, not months.<\/p>\n\n\n\n<p>This is the story of how Gilles did it, what he learned along the way, and what it means for anyone with a business idea and zero engineering experience.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"305\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4-1024x305.avif\" alt=\"\" class=\"wp-image-4702\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4-1024x305.avif 1024w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4-300x89.avif 300w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4-768x229.avif 768w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4-1536x458.avif 1536w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-4.avif 1751w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Masters_of_Photography_the_business_behind_the_builder\"><\/span><strong>Masters of Photography: the business behind the builder<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>To understand why Pixwise exists, you need to understand Gilles&#8217; world. He co-founded <a href=\"https:\/\/www.mastersof.photography\/en\" rel=\"nofollow noopener\" target=\"_blank\">Masters of Photography<\/a> in 2018 alongside filmmaker Chris Ryan. The platform is an online premium education business offering in-depth video masterclasses with some of the greatest photographers alive. Not webinars or quick tips, but proper multi-hour courses filmed on location around the world, following each top photographer as they work.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"340\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8-1024x340.avif\" alt=\"\" class=\"wp-image-4706\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8-1024x340.avif 1024w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8-300x100.avif 300w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8-768x255.avif 768w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8-1536x510.avif 1536w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-8.avif 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p>The roster is remarkable. Steve McCurry, the photographer behind the iconic Afghan Girl portrait. Joel Meyerowitz, the legendary street and art photographer who has been working for over 50 years. Albert Watson, one of the most influential portrait and fashion photographers in the world, filmed in his New York studio and on location in Morocco. David Yarrow, the bestselling fine-art photographer, filmed in Montana and Kenya. Paul Nicklen and Cristina Mittermeier, both renowned conservation photographers, filmed in British Columbia and Mexico respectively. Nick Danziger, the celebrated photojournalist, filmed in Armenia, France, and the UK. Each masterclass runs between four and eight hours, split into episodes of 15 to 20 minutes, and together they form a genuinely unique library of photographic education. The platform has attracted more than 30,000 photographers.<\/p>\n\n\n\n<p>Masters of Photography is very much alive and growing. They recently launched a new book with Paul Nicklen, one of their photographers, and they continue to release new content and host live Q&#038;A sessions with the masters.&nbsp;<\/p>\n\n\n\n<p>But Gilles had been thinking for a long time about a gap in what they offered.<\/p>\n\n\n\n<p>The masterclasses are designed for photographers who already have a solid foundation. They teach you to develop your eye, find your own style, understand the philosophy and creative process behind great photography. They are, deliberately, not technical courses about camera settings or the physics of light.<\/p>\n\n\n\n<p>&#8220;The masterclasses expect a certain level of mastery already,&#8221; Gilles explains. &#8220;They are about developing your eye, finding your voice. But what about the person who is just starting out? The person who does not yet understand f-stops and shutter speeds and the theory of light? There was always this gap between what we offered and the more basic technical needs of a beginner photographer.&#8221;<\/p>\n\n\n\n<p>The challenge was how to fill that gap. Teaching camera settings and technical fundamentals through traditional video is, as Gilles puts it, &#8220;very dry.&#8221; Anyone who has tried to learn aperture values and shutter speed relationships from a YouTube video knows what he means. It is information-dense, context-dependent, and the moment you are standing in a field at golden hour trying to photograph a landscape, you have forgotten half of what you watched.<\/p>\n\n\n\n<p>Gilles wanted something different. Something interactive. Something that could bridge the gap between sitting at home watching a lesson and standing in the field trying to make a photograph. Something that felt less like a textbook and more like having a knowledgeable friend with you.<\/p>\n\n\n\n<p>That is when he started thinking about AI.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"The_idea_behind_Pixwise\"><\/span><strong>The idea behind Pixwise<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>The concept is elegant. Pixwise Studio does two things, and both of them exploit what AI is genuinely good at: processing information quickly and giving structured, personalised feedback.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"538\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-5-1024x538.avif\" alt=\"\" class=\"wp-image-4703\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-5-1024x538.avif 1024w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-5-300x158.avif 300w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-5-768x403.avif 768w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-5.avif 1200w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p>The first feature is an AI-powered camera settings advisor. Before you go out to shoot, you tell Pixwise what you want to photograph. A landscape. A portrait. A street scene at night. It asks you a series of questions about the conditions, what you are trying to achieve, what equipment you have. Those answers form a prompt that gets sent to an LLM, and it comes back with structured, practical advice: recommended camera settings, tips on approach, things to watch out for.<\/p>\n\n\n\n<p>The second feature is where it gets really interesting. After you have taken your photographs, you can upload an image and Pixwise gives you a detailed, structured critique. Composition, exposure, technique. It tells you what works, what does not, what you could improve if you were to reshoot, and what you could fix in post-production. It is like having a professional photographer look over your shoulder and give you honest, specific feedback.<\/p>\n\n\n\n<p>The photo critique feature uses Google&#8217;s Gemini Flash model, which Gilles chose because it is particularly strong at analysing images and because it had a generous free token allowance when he was building. The camera settings advisor runs through Claude&#8217;s API. Both return structured responses so the feedback is consistent, not just a wall of text that changes format every time.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"576\" height=\"1024\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7-576x1024.avif\" alt=\"\" class=\"wp-image-4705\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7-576x1024.avif 576w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7-169x300.avif 169w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7-768x1365.avif 768w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7-864x1536.avif 864w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-7.avif 1080w\" sizes=\"auto, (max-width: 576px) 100vw, 576px\" \/><\/figure>\n<\/div>\n\n\n<p>&#8220;I thought, okay, this is quite a nice tool,&#8221; Gilles says. &#8220;The question is, is this going to work? Is this of interest? It is very difficult to know. And if you had to build this from scratch a few years ago, it would have cost a fair bit of money with actually no guarantee of success. It is a big punt.&#8221;<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"1024\" src=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6-1024x1024.avif\" alt=\"\" class=\"wp-image-4704\" srcset=\"https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6-1024x1024.avif 1024w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6-300x300.avif 300w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6-150x150.avif 150w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6-768x768.avif 768w, https:\/\/agentaya.com\/wp-content\/uploads\/2026\/04\/image-6.avif 1080w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p>A few years ago, &#8220;a big punt&#8221; would have meant writing a cheque for tens of thousands of pounds to an agency, waiting months, and hoping for the best. That is not a punt most small business owners can afford to take. But this time, Gilles decided to build it himself.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Why_this_matters_beyond_photography\"><\/span><strong>Why this matters beyond photography<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Before we get into the details of how Gilles built Pixwise, it is worth stepping back and understanding why this story matters. Not just for photographers, not just for education businesses, but for anyone who runs or wants to start a small business.<\/p>\n\n\n\n<p>For decades, the economics of building digital products have been brutally simple. If you have an idea for a piece of software, a web application, an online tool, you have three options. You can learn to code yourself, which takes months or years of serious study. You can hire developers, which costs tens of thousands at minimum and comes with all the risks of managing a technical project you do not fully understand. Or you can use no-code tools like Bubble or Webflow, which are easier but lock you into their platforms and come with real limitations.<\/p>\n\n\n\n<p>Vibe coding, the term coined by Andrej Karpathy (co-founder of OpenAI and former AI director at Tesla) in February 2025, changes the equation entirely. The basic idea is that you describe what you want in plain language, and AI generates the actual code. You focus on the product, the problem, the user experience. The AI handles the syntax.<\/p>\n\n\n\n<p>In February 2025, Karpathy posted: &#8220;There is a new kind of coding I call vibe coding, where you fully give in to the vibes, embrace exponentials, and forget that the code even exists.&#8221; Collins Dictionary named it their Word of the Year for 2025. By early 2026, it has gone from a tweet to a movement. According to community analyses, 63% of active vibe coding users are not developers. They are founders, product managers, marketers, and business owners building real, functional products. Y Combinator reported that a quarter of startups in its Winter 2025 batch had codebases that were 95% AI-generated.<\/p>\n\n\n\n<p>The tools that make this possible (Cursor, Lovable, Replit, Bolt, Claude Code, and others) have matured to the point where a non-technical person can go from an idea to a deployed web application in days. The cost is typically a handful of AI subscriptions. The code is real, exportable, deployable anywhere. It is not a drag-and-drop website builder with a ceiling.<\/p>\n\n\n\n<p>But, and Gilles&#8217; experience illustrates this clearly, there is a meaningful difference between generating code quickly and building a product well. That difference is about structure, process, and knowing when to iterate and when to ship.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Starting_with_Lovable_and_learning_its_limits\"><\/span><strong>Starting with Lovable, and learning its limits<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Like many people who start exploring vibe coding, Gilles began with <a href=\"https:\/\/lovable.dev\/\" rel=\"nofollow noopener\" target=\"_blank\">Lovable<\/a>, one of the most popular &#8220;prompt to product&#8221; platforms. The appeal is obvious. You describe your idea, and Lovable generates a working application that looks polished and professional, often within minutes.<\/p>\n\n\n\n<p>&#8220;Lovable was very good at creating something that looks great,&#8221; Gilles says. &#8220;But I think the trap I was falling into is that these tools tell you: just give us your idea and we will build it. So you come with something very broad, put it in the machine, and it builds something. But you realise that once it is built, it is very difficult to change it and move it in the direction you actually want. Quite often it breaks, and then it goes into some sort of a loop.&#8221;<\/p>\n\n\n\n<p>This is a pattern that almost every non-technical vibe coder runs into at some point. The initial result is impressive, even magical. But software is not a single moment of creation. It is an ongoing process of refinement, adjustment, and debugging. When you have generated a large codebase in one go and something breaks, figuring out where the problem is and how to fix it without breaking something else is genuinely hard, especially if you do not understand the code.<\/p>\n\n\n\n<p>The token economics make this worse. Free tiers give you just enough to get excited, but serious iteration burns through tokens fast. &#8220;All right, you need to change this. Oh, now all your tokens are gone. You have to upgrade, and wait, I just used one prompt,&#8221; Gilles recalls.<\/p>\n\n\n\n<p>He did not give up on vibe coding. But he recognised that he needed a more structured method, something that gave him control over the process rather than handing the entire project to an AI and hoping for the best.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Finding_a_method_that_works\"><\/span><strong>Finding a method that works<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>The turning point came through <a href=\"https:\/\/www.lennysnewsletter.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Lenny&#8217;s Newsletter<\/a>, one of the most respected product management publications. Gilles found an episode featuring <a href=\"https:\/\/www.linkedin.com\/in\/zev-arnovitz\" rel=\"nofollow noopener\" target=\"_blank\">Zevi Arnovitz<\/a>, a product manager at Meta who, despite having no technical background, had figured out how to build and ship real products using AI tools. His engineering team at Meta actually asks him to teach them how he does what he does.<\/p>\n\n\n\n<p>The episode, <a href=\"https:\/\/www.lennysnewsletter.com\/p\/the-non-technical-pms-guide-to-building-with-cursor\" rel=\"nofollow noopener\" target=\"_blank\">The non-technical PM&#8217;s guide to building with Cursor<\/a>, lays out a complete workflow built around a series of structured prompts. The approach is fundamentally different from what Lovable and similar tools offer. Instead of dumping your entire vision into a single prompt and letting the AI run, Arnovitz breaks the process into deliberate, sequential steps.<\/p>\n\n\n\n<p>The workflow goes something like this. You start by telling Claude what its role is and what tools you will be using to build the product. You specify how you want Claude to communicate with you. You provide a structured workflow for Claude to follow. Then, for each individual feature or bug fix, you brainstorm the requirement, Claude asks you clarifying questions to make sure it understands the brief correctly, and then Claude creates a specific prompt. You take that prompt and paste it into Cursor, the AI-powered code editor, which generates the actual code. Cursor gives you a status report, which you feed back to Claude for review. The two tools check each other&#8217;s work.<\/p>\n\n\n\n<p>This dual-tool approach is more involved than &#8220;just tell the AI what you want.&#8221; But that additional structure is exactly what makes it work for serious projects. You are not generating an entire application in one shot and then trying to debug it. You are building feature by feature, prompt by prompt, with two AI systems collaborating and reviewing each other&#8217;s output.<\/p>\n\n\n\n<p>&#8220;What is amazing,&#8221; Gilles says, &#8220;is that by working this way and asking Claude to be your partner, Claude suggests ideas, suggests structure, and then you work together. It is like having a senior developer sitting next to you and implementing everything.&#8221;<\/p>\n\n\n\n<p>The word &#8220;partner&#8221; comes up repeatedly when Gilles talks about the process. Not a tool, not an assistant, but a partner. He describes conversations with Claude where he would explain what he was trying to achieve, and Claude would come back with suggestions he had not considered, structural improvements, alternative approaches. The AI was not just executing instructions, it was genuinely contributing to the product strategy.<\/p>\n\n\n\n<p>This is something that people who have not used these tools in a sustained way often do not grasp. The conversation between a human and an AI during a build process is not just a series of commands and outputs. It is truly collaborative: you bring the domain knowledge, the understanding of the user, the sense of what the product should feel like. And the AI brings the ability to structure, implement, and suggest based on patterns it has seen across millions of projects. The combination is more powerful than either alone.<\/p>\n\n\n\n<p>One practical tip Gilles mentioned: when you are spending hours giving instructions to Claude and Cursor, typing sometimes becomes a genuine bottleneck. Voice-to-text tools like <a href=\"https:\/\/wisprflow.ai\" rel=\"nofollow noopener\" target=\"_blank\">Wispr Flow<\/a> can speed things up significantly. It is a detail that Karpathy himself flagged in his original vibe coding post, and Arnovitz also uses it in his workflow. When your primary interface with a development tool is natural language, being able to speak rather than type is a surprisingly meaningful productivity gain.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"What_Gilles_actually_built_with_and_what_it_cost\"><\/span><strong>What Gilles actually built with, and what it cost<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>For anyone considering a similar project, here is the full picture of what Gilles used.<\/p>\n\n\n\n<p>He used <a href=\"https:\/\/claude.ai\/\" rel=\"nofollow noopener\" target=\"_blank\">Claude<\/a> on the Pro plan at \u00a315 per month as his primary AI partner for everything: brainstorming, product planning, prompt creation, code review, SEO strategy, pricing strategy, and general business thinking. <a href=\"https:\/\/cursor.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Cursor<\/a> served as the AI-powered code editor where the actual code got written, and Gilles was able to stay on the free tier for this.<\/p>\n\n\n\n<p>For the AI features that power Pixwise itself (as opposed to the AI tools used to build it), Gilles integrated the Claude API for the camera settings advisor and Google Gemini Flash for the photo critique feature. He chose Gemini Flash specifically because of its strong image analysis capabilities and its generous token allowance before hitting paid tiers.<\/p>\n\n\n\n<p>The frontend is React with TypeScript, built using Vite. The backend runs on <a href=\"https:\/\/supabase.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Supabase<\/a>, an open-source alternative to Firebase, which handles the database, authentication, and Google social login. Deployment goes through GitHub to <a href=\"https:\/\/vercel.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Vercel<\/a>. Transactional emails (welcome emails, notifications, and so on) are handled by <a href=\"https:\/\/resend.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Resend<\/a>.<\/p>\n\n\n\n<p>Gilles also tried to implement Apple social login alongside Google, but found it significantly more complex to set up in this kind of build. He wisely abandoned it rather than sinking more time into a feature that was not essential.<\/p>\n\n\n\n<p>When asked about the total cost, Gilles is almost embarrassed by how low it is. &#8220;It costs virtually nothing. I think it might have cost me $100.&#8221; That covers the Claude Pro subscription, and a handful of other small subscriptions.<\/p>\n\n\n\n<p>$100. For a fully functioning web application with user authentication, payment processing, two separate AI integrations, a content management system, and a deployment pipeline.<\/p>\n\n\n\n<p>To put that in perspective, Gilles has years of experience working with agencies and development teams, both in the UK and with distributed teams in India, Berlin and Eastern Europe. He knows what these things cost. When asked what a comparable product would have cost to build traditionally in 2018, when he started Masters of Photography, he did not hesitate: &#8220;$25,000 to $40,000 for a first version. If you want a high-quality one, it could easily be double that.&#8221;<\/p>\n\n\n\n<p>That is a cost reduction of somewhere between 99.6% and 99.8%. And the traditional approach would have taken months, not weeks, with no guarantee the concept would resonate with users.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Three_weeks_a_hundred_hours_and_the_%E2%80%9Caddiction%E2%80%9D_of_building\"><\/span><strong>Three weeks, a hundred hours, and the &#8220;addiction&#8221; of building<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Gilles estimates the full build took three to four weeks and around 100 hours. That includes everything: learning the tools and the Arnovitz workflow, registering for and connecting all the various services (Supabase, Vercel, Resend, Google backend, and the various APIs), building the core product, iterating on design, and adding features that were not part of the original plan.<\/p>\n\n\n\n<p>A significant chunk of that time, and this is a point Gilles emphasises, was spent on setup and learning. Registering accounts, connecting APIs, figuring out how all the pieces fit together. &#8220;There are a lot of little pieces, APIs that you need to connect from one place to another. That takes a while.&#8221; For someone doing this for the second time, that overhead would largely disappear.<\/p>\n\n\n\n<p>There is also the question of scope creep, though &#8220;scope creep&#8221; sounds too negative for what Gilles describes. It is more like scope enthusiasm.<\/p>\n\n\n\n<p>&#8220;One of the challenges is that because you realise how easy it is to create stuff, you never stop,&#8221; he says, laughing. &#8220;You think, oh my God, I could do this, I could add that, I am not happy with this. And it goes on and on and on.&#8221;<\/p>\n\n\n\n<p>Features that emerged from this process include a learning journal where users can track their progression over time, a challenges feature that sets specific photography tasks to help users practice targeted skills, and a community section where photographers can share their work and give each other feedback. None of these were in the original plan. They came from the fact that building them was, suddenly, within reach.<\/p>\n\n\n\n<p>&#8220;It probably could have been built much faster,&#8221; Gilles admits. &#8220;But there was always something I wanted to add.&#8221;<\/p>\n\n\n\n<p>When asked what it would take to build a comparable product today, knowing everything he now knows and without the learning curve, Gilles thinks for a moment. &#8220;You can put something out there in maybe 20 hours. It is after that, the finessing, the finishing of the brand, making sure you are happy with everything, that can take a longer time.&#8221;<\/p>\n\n\n\n<p>Twenty hours from idea to functional product. That is the number that should make every business owner sit up and pay attention. Not because every idea should be built in 20 hours, but because the cost of testing an idea has fallen so dramatically that the entire calculus of entrepreneurship has changed.<\/p>\n\n\n\n<p>In the past, testing a digital product idea meant a serious financial commitment. $25,000 to $80,000. Months of development. You had better be pretty sure it was going to work. Now, the cost of finding out whether an idea has legs is a few weekends and $100. You can afford to be wrong. You can afford to try three or four ideas before one sticks. As Gilles puts it: &#8220;If I have another idea tomorrow, I will do another one. And way faster. You can try more things until you find something that captures the imagination. Which you could not do in the past.&#8221;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Where_it_was_not_smooth_an_honest_look_at_the_friction\"><\/span><strong>Where it was not smooth: an honest look at the friction<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>One of the things we at AgentAya care about is telling the real story, not the polished version. Gilles was refreshingly honest about where the process was frustrating, and his experience aligns with what we hear from many non-technical builders.<\/p>\n\n\n\n<p>Deployment was a recurring source of frustration. The code gets written in Cursor, pushed to GitHub, and then deployed through Vercel. That pipeline is standard and generally works well, but when it does not, tracking down the source of the error can be painful. &#8220;You try to find the errors in the console, send them back to Claude, Claude suggests some fixes, and it can go on for a bit,&#8221; Gilles says. When you do not fully understand the code, diagnosing why a deployment is failing requires a different kind of patience than writing the code in the first place.<\/p>\n\n\n\n<p>Design is another area where Gilles feels the current tools fall short. Claude and Cursor are strong at generating functional code, but visual design, branding, colour schemes, and the overall aesthetic feel of a product are harder to get right through prompts alone. &#8220;The one thing that is clearly not that great yet, is design,&#8221; Gilles says. &#8220;I tried different iterations of brands and colour schemes and I am still not convinced.&#8221; The Pixwise logo, for example, was generated by Claude. It works, but Gilles is the first to say it is not where he wants it to be.<\/p>\n\n\n\n<p>He has heard that some vibe coders are using Lovable specifically for design, creating the visual look of pages in Lovable where the UI generation is stronger, and then porting those designs to Cursor for implementation. He is considering trying this hybrid approach.<\/p>\n\n\n\n<p>The SEO content that Claude generated for Pixwise also comes with a caveat. &#8220;It is probably a bit AI-ish, so not very deep in terms of content,&#8221; Gilles acknowledges. &#8220;But the idea was to create something basic to start with and then see where it goes.&#8221; This is an honest assessment that many people generating AI content would do well to make. AI-generated content can get you started, but if you want content that truly resonates with a knowledgeable audience, human editing and domain expertise still matter.<\/p>\n\n\n\n<p>These are real limitations, and they are worth knowing about before you start. Vibe coding does not produce perfect results. It produces a first version, sometimes a very good first version, that then needs iteration, refinement, and, in some areas, the input of someone with specialist skills. But the crucial point is that it gets you from zero to something real, something you can test, show to users, and iterate on. The gap between &#8220;I have an idea&#8221; and &#8220;I have a product&#8221; has shrunk from months and tens of thousands of pounds to weeks and a hundred dollars.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Building_step_by_step_the_single_most_important_lesson\"><\/span><strong>Building step by step: the single most important lesson<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>If there is one piece of advice that Gilles would emphasise above all others, it is this: build incrementally.<\/p>\n\n\n\n<p>&#8220;I think it is very, very important to work step by step,&#8221; he says, and there is real emphasis in his voice. &#8220;This is where the difficulty with Lovable was. You just give your project and it goes and builds something. But I think you need to catch bugs when they appear, straight away.&#8221;<\/p>\n\n\n\n<p>The metaphor he uses is walking versus running. Tools like Lovable start running while you are still learning to walk. They generate a lot of code quickly, which feels incredible, but if something goes wrong in that mass of generated code, you are lost. You do not know where the problem is, you do not understand the code well enough to diagnose it, and the AI&#8217;s attempts to fix it can create new problems.<\/p>\n\n\n\n<p>The Arnovitz method that Gilles followed avoids this by treating each feature as a discrete unit. You build one thing, you test it, you make sure it works, and then you move to the next thing. If something breaks, you know exactly which change caused it. If you need to debug, you are dealing with a small, contained piece of code, not an entire application.<\/p>\n\n\n\n<p>This is not a new insight in software development. Professional engineers have been preaching incremental development for decades. But it is particularly important in the vibe coding context because the temptation to generate everything at once is so strong. The tools make it feel like you should be able to describe your entire product and have it appear. And sometimes that works for simple projects. But for anything with real complexity, payment systems, user accounts, API integrations, multiple AI models, the step-by-step approach is not just better, it is essentially mandatory.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"AI_as_more_than_a_builder\"><\/span><strong>AI as more than a builder<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>One of the most interesting aspects of Gilles&#8217; experience is how far beyond code generation he pushed the AI tools. Claude was not just the developer. It was the business strategist, the SEO planner, the pricing consultant, and the product advisor.<\/p>\n\n\n\n<p>The pricing page on Pixwise, with its tiered structure of Starter, Hobbyist, and Studio plans, was developed collaboratively with Claude. Gilles would give Claude a different system prompt, essentially saying &#8220;you are now my business manager and pricing strategist,&#8221; and work through questions like: How many free sessions should we offer? What makes the premium tier feel genuinely premium? Should we include PDF exports of feedback, an analytics dashboard, challenge history? These are business decisions, not technical ones, and they are exactly the kind of structured thinking where AI can be a surprisingly good sparring partner.<\/p>\n\n\n\n<p>&#8220;You can give Claude a different prompt at the beginning and say, right, you are my business manager strategist. How should I build this?&#8221; Gilles explains. &#8220;And then suddenly you get new ideas. It helps you structure the whole programme.&#8221;<\/p>\n\n\n\n<p>He used the same approach for SEO. Claude acting as an SEO specialist produced a content plan, keyword strategy, and then generated the initial content for the Pixwise blog. Gilles used Claude&#8217;s advice on tool selection too, discovering services like Resend for transactional emails through conversations about what infrastructure he needed.<\/p>\n\n\n\n<p>This is something that gets lost in the conversation about vibe coding. The headlines focus on &#8220;non-technical person builds an app,&#8221; which is impressive enough. But the real story is broader. A non-technical person, with the right AI tools, can now handle the full spectrum of building and launching a product: the strategy, the product design, the development, the marketing infrastructure, the pricing, the SEO, the content. Not all of it will be perfect. Some of it will need refinement by specialists. But you can get a credible first version of all of it done by yourself, and then invest selectively in the areas that need the most improvement.<\/p>\n\n\n\n<p>For a small business owner or solo entrepreneur, that changes everything. You are no longer dependent on a developer to build, a marketing agency to plan your SEO, a consultant to help with pricing, and a designer to make it look good. You can do a reasonable version of all of it, identify which parts need professional help, and then spend your money where it matters most.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"What_Gilles_would_tell_another_entrepreneur\"><\/span><strong>What Gilles would tell another entrepreneur<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>When asked what advice he would give to someone inspired by his story, Gilles is thoughtful about it. He does not leap to the technology. He starts with the fundamentals.<\/p>\n\n\n\n<p>&#8220;The important bit now is to come up with a good idea,&#8221; he says. &#8220;It is the obvious stuff. What is the problem you are trying to solve? Define the problem properly. And then from the problem, try to come up with a product that is structured in a way that answers it.&#8221;<\/p>\n\n\n\n<p>This might sound basic, but it is more important now than ever. When the cost of building is near zero, the bottleneck shifts entirely to the quality of the idea and the clarity of your thinking about the problem. In the old world, many bad ideas never got built because the cost of building was itself a filter. Now, anyone can build anything. The question is whether anyone wants it.<\/p>\n\n\n\n<p>Gilles&#8217; second piece of advice comes from his own mistake. He built too much before testing.<\/p>\n\n\n\n<p>&#8220;I think what I should have done is build something very quickly, in a week or two weeks, and then send it out for testing first. You are so excited, and you build the whole thing before you ask for feedback.&#8221;<\/p>\n\n\n\n<p>He is now in a testing phase with 10 to 15 photographers using Pixwise and providing feedback. The features they are asking for, the learning journal, the challenges system, the community section, are things Gilles had not originally planned. &#8220;You always have one idea about what you think your product should be. And then you realise there is a lot of good advice and good ideas that you have not thought about.&#8221;<\/p>\n\n\n\n<p>The third point is about distribution, and it is the one that tempers the excitement. AI can help you build a product. It can help with SEO and content. It can even help you think through your go-to-market strategy. But it will not magically bring you thousands of paying customers.<\/p>\n\n\n\n<p>&#8220;The very difficult bit is how you distribute your product,&#8221; Gilles says.<\/p>\n\n\n\n<p>He posted about Pixwise on LinkedIn when he launched, but his network spans broadcasting, gaming, and photography, and is not concentrated enough in any one area for the &#8220;building in public&#8221; approach to generate meaningful traction. He has plans to target photography clubs directly, using Claude to help build scraping tools that can identify and collect contact information for photography clubs across the UK and US, creating a database for outreach. But this is work that requires effort, persistence, and experimentation. AI helps with the mechanics, but you still need a strategy and the willingness to execute it.<\/p>\n\n\n\n<p>On maintenance, Gilles says there is virtually nothing to do on an ongoing basis. The product runs itself. The main thing to watch is the token consumption on the AI APIs as usage grows, and the possibility that database costs on Supabase will need to step up to a paid tier as the number of users increases. But both of these are problems that only arise if the product is succeeding, which makes them good problems to have.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"The_bigger_picture_what_this_means_for_small_business_owners\"><\/span><strong>The bigger picture: what this means for small business owners<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>There is a tendency, when talking about vibe coding and AI-powered development, to frame it as a Silicon Valley phenomenon. Y Combinator startups with 95% AI-generated codebases. Technical founders moving faster. Engineers becoming more productive.<\/p>\n\n\n\n<p>Gilles&#8217; story is important precisely because it is none of those things. He is a business owner in his fifties, based in England, with a background in broadcasting, gaming, and digital marketing. He runs an education platform with a small team. He has worked with agencies in Europe and distributed development teams in India and Eastern Europe. He has lived through the previous waves of disruption in how digital products get built: the shift from expensive European and American agencies to lower-cost talent in Asia, the rise of no-code tools, the constant churn of platforms and frameworks.<\/p>\n\n\n\n<p>This wave is different. Not because AI is smarter than those previous shifts, though it is, but because of what it does to the economics of experimentation. The old model was: have an idea, spend $50,000 and three months, find out if it works. The new model is: have an idea, spend $100 and three weeks, find out if it works. And if it does not work, try another idea next month.<\/p>\n\n\n\n<p>For small business owners in particular, for the supermarket owner who wants to automate daily reports, the photographer who wants an AI learning tool, the consultant who needs a client portal, the retailer who wants a custom inventory system, this is transformative. Not because every one of these ideas will succeed, but because the cost of trying has dropped below the threshold where failure matters. You can experiment. You can iterate. You can discover what works by building it, not by commissioning a feasibility study.<\/p>\n\n\n\n<p>The technology still has real limitations. Design is not where it needs to be. Security requires attention, with research indicating that a meaningful percentage of AI-generated code contains vulnerabilities. The &#8220;never-ending build&#8221; that Gilles describes, where the ease of adding features makes it hard to stop and ship, is a real trap. And distribution, finding and converting actual paying customers, remains stubbornly resistant to automation.<\/p>\n\n\n\n<p>But the trajectory is unmistakable. The tools are improving rapidly. Costs are falling. The community of non-technical builders is growing and sharing what they learn. As Gilles puts it, with the directness of someone who has just lived through the experience: &#8220;There is no excuse, really. If you have an idea and you want to do something, it is not that you cannot because you do not have money.&#8221;<\/p>\n\n\n\n<p>The barrier to building a digital product is no longer financial or technical. It is about having a clear idea, the willingness to learn a new process, and the discipline to build step by step. Everything else, the AI can help with.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Resources_and_links\"><\/span><strong>Resources and links<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p><strong>Gilles&#8217; projects:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.pixwise.studio\/\" rel=\"nofollow noopener\" target=\"_blank\">Pixwise Studio<\/a>: the AI photography learning platform<\/li>\n\n\n\n<li><a href=\"https:\/\/www.mastersof.photography\/en\" rel=\"nofollow noopener\" target=\"_blank\">Masters of Photography<\/a>: online masterclasses with world-renowned photographers<\/li>\n\n\n\n<li><a href=\"https:\/\/www.linkedin.com\/in\/gillesstorme\/\" rel=\"nofollow noopener\" target=\"_blank\">Gilles Storme on LinkedIn<\/a><\/li>\n<\/ul>\n\n\n\n<p><strong>The workflow Gilles followed:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.lennysnewsletter.com\/p\/the-non-technical-pms-guide-to-building-with-cursor\" rel=\"nofollow noopener\" target=\"_blank\">The non-technical PM&#8217;s guide to building with Cursor<\/a> by Zevi Arnovitz on Lenny&#8217;s Newsletter<\/li>\n<\/ul>\n\n\n\n<p><strong>Tools used to build Pixwise:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/claude.ai\/\" rel=\"nofollow noopener\" target=\"_blank\">Claude<\/a> for planning, code review, strategy, and content<\/li>\n\n\n\n<li><a href=\"https:\/\/cursor.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Cursor<\/a> as the AI-powered code editor<\/li>\n\n\n\n<li><a href=\"https:\/\/supabase.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Supabase<\/a> for backend and authentication<\/li>\n\n\n\n<li><a href=\"https:\/\/vercel.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Vercel<\/a> for deployment<\/li>\n\n\n\n<li><a href=\"https:\/\/resend.com\/\" rel=\"nofollow noopener\" target=\"_blank\">Resend<\/a> for transactional email<\/li>\n<\/ul>\n\n\n\n<p><strong>Beginner-friendly vibe coding platforms:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/agentaya.com\/ai-review\/lovable\/\" data-type=\"ai_tool\" data-id=\"2166\">Lovable<\/a>: fast prototyping, strong UI generation<\/li>\n\n\n\n<li><a href=\"https:\/\/agentaya.com\/ai-review\/bolt-new\/\" data-type=\"ai_tool\" data-id=\"4629\">Bolt<\/a>: quick prototyping with in-browser preview<\/li>\n\n\n\n<li><a href=\"https:\/\/agentaya.com\/ai-review\/replit\/\" data-type=\"ai_tool\" data-id=\"2190\">Replit<\/a>: browser-based development with AI agent<\/li>\n\n\n\n<li><a href=\"https:\/\/v0.app\/\" rel=\"nofollow noopener\" target=\"_blank\">v0 by Vercel<\/a>: UI component generation<\/li>\n<\/ul>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p><em>This interview was conducted by <a href=\"https:\/\/agentaya.com\" data-type=\"page\" data-id=\"6\">AgentAya<\/a> as part of our ongoing series exploring how real business owners are using AI tools to build, grow, and transform their businesses. Have a story to share? We would love to hear from you.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>An AgentAya interview with Gilles Storme, co-founder of Masters of Photography and creator of Pixwise Studio. Gilles Storme has spent&#8230;<\/p>\n","protected":false},"author":2,"featured_media":4706,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_kad_post_transparent":"","_kad_post_title":"","_kad_post_layout":"","_kad_post_sidebar_id":"","_kad_post_content_style":"","_kad_post_vertical_padding":"","_kad_post_feature":"","_kad_post_feature_position":"","_kad_post_header":false,"_kad_post_footer":false,"_kad_post_classname":"","footnotes":""},"categories":[79],"tags":[],"class_list":["post-4700","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-case-study"],"acf":[],"_links":{"self":[{"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/posts\/4700","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/comments?post=4700"}],"version-history":[{"count":2,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/posts\/4700\/revisions"}],"predecessor-version":[{"id":4708,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/posts\/4700\/revisions\/4708"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/media\/4706"}],"wp:attachment":[{"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/media?parent=4700"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/categories?post=4700"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/agentaya.com\/nl\/wp-json\/wp\/v2\/tags?post=4700"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}