By Abhi, Founder of Lowvyn

Every time I share Lowvyn on a major social platform, someone calls it AI slop.

I understand the skepticism. The internet is currently saturated with AI-generated products that appear complete but fail in practical use. Many tools are designed to capture traffic rather than solve real problems, so the doubt is justified.

I want to clarify what it means to build a genuine product with AI, as current discussions often overlook an important distinction. Using AI as a shortcut to create a superficial product is very different from using it to accelerate the development of something you understand deeply. Lowvyn represents the latter, and I believe it is important to explain why.

The Problem Came First

Before there was any code, there was a spreadsheet.

I built Lowvyn while preparing for my first child. The most stressful purchase was a Graco 4Ever car seat. A reliable car seat is essential before bringing a baby home, and the high price means poor timing can be costly.

To address this, I tracked prices for nearly two months, from Black Friday through Boxing Day, manually checking Amazon.ca, Walmart.ca, Best Buy Canada, Canadian Tire, and Toys R Us Canada. The price remained at $449 across all retailers, with no significant discounts despite extensive holiday promotions. On January 4th, we purchased it for $449 on Amazon.

On January 31st, Amazon reduced the price by $100, offering the same car seat for $349.

I contacted Amazon support, and after a thirty-minute conversation, they credited $100 due to a price-matching policy for purchases made within 30 days. We received the refund only because I had the purchase details ready and was willing to spend the time advocating for it.

This experience made it clear that price history is valuable both before and after a purchase. Knowing past prices helps determine if a sale is genuine and whether you paid a fair amount. If a retailer lowers the price within a price match window, having this information allows you to act quickly and confidently.

This is the tool I needed, and it is what Lowvyn provides.

This origin is important not only as a personal story but also as a technical foundation. I approached this problem with domain knowledge that AI alone cannot provide. I understood the nuances of Canadian retail pricing across multiple retailers, the differences in promotions, and the timing of genuine discounts. My experience informed what a real solution needed to address.

What AI Helped With

I used Claude Code extensively to build Lowvyn, and I want to be fully transparent about this.

Claude Code helped me write and iterate on backend logic faster than I could have alone. It helped me structure components, debug edge cases, and move through implementation at a pace that simply would not have been possible for a solo founder working alongside other employment commitments.

Without AI assistance, Lowvyn, in its current form, covering Amazon.ca, Walmart.ca, and Best Buy Canada with daily price updates across thousands of products, would have taken years instead of months. That acceleration is real, and I am not going to pretend otherwise.

However, that statement does not capture the full picture.

What AI Could Not Do

AI could not identify the unique patterns and quirks of Canadian retail pricing compared to the American market. It did not recognize that inflated discounts and list prices are especially common in Canadian baby gear retail. AI also did not determine that a Lowvyn Score from 0 to 100 was the appropriate metric, nor what it should measure and prioritize.

AI did not select baby products as the initial category. That decision was based on personal experience, months of research, and an understanding that parents making these purchases are often short on time, financially constrained, and highly motivated to make informed choices. This is a user insight, not something generated by a prompt.

AI did not determine what Lowvyn should not be. The terms of service make it clear: Lowvyn is not a deal aggregator or coupon site. It is a price-tracking tool that empowers users to make informed decisions. This approach was a deliberate choice to respect user intelligence, not a decision made by AI.

Starting Small on Purpose

Launching with baby products was a strategic decision, not simply the easiest option.

Focusing on a single category allowed me to validate the model before expanding. I was building for a user I understood well—myself a year ago. This ensured the product was genuinely useful to a real audience before broadening its scope.

Canadian parents shopping for baby gear face a fragmented retail landscape. Amazon.ca, Walmart.ca, and Best Buy Canada are primary sources for daily price tracking, while Canadian Tire, Costco Canada, Toys R Us Canada, and Hudson’s Bay serve specific needs and seasonal sales. Each retailer has unique promotion schedules, pricing conventions, and price match policies. Understanding this landscape, rather than simply collecting data, is what makes Lowvyn valuable.

This disciplined approach contrasts with building features for the sake of volume. Lowvyn began with a narrow focus, remained so until the core experience was solid, and is now expanding because the foundation can support growth.

The same discipline applies to cost management. Lowvyn is free, which required building scalable infrastructure and making careful decisions about what to track, how frequently, and at what level of detail. These choices have long-term impacts. AI can execute instructions, but it does not make these strategic decisions.

Transparency Is Not a Feature. It Is a Foundation.

I want to address something that is often overlooked in products like this.

Lowvyn uses affiliate links. When you click a product and make a purchase on Amazon.ca, we may earn a small commission. This allows the tool to remain free without charging users or selling data. Our terms of service make it clear that affiliate relationships do not influence product selection, rankings, or deal assessments. Scores are based solely on price history data.

We do not sell personal information or require an account to use core features. If you opt in for price alerts, we store your email address, and nothing more. The privacy policy is written in clear language because there is nothing to hide.

I am disclosing AI usage for the same reason: to build trust with Lowvyn users by being transparent about how it was built, what data it uses, and its purpose.

Why This Post Exists

I am sharing this because Lowvyn is both a community project and a product.

I am a solo founder building something I believe is genuinely useful for Canadian shoppers. I use AI tools to accelerate development, but I also bring domain knowledge, product judgment, and a personal commitment to its success. I am transparent about all aspects, including limitations, categories not yet covered, and ongoing improvements.

If you have found Lowvyn useful, sharing it with others who may benefit is more valuable than any marketing campaign. If you notice something broken or missing, your feedback directly shapes future development. This feedback loop is essential.

The key question is not whether AI contributed to building Lowvyn, but whether a real person with sound judgment is guiding it, whether the tool is effective, and whether it serves your interests.

I believe the answer to all three is yes, and I am building transparently so you can hold me accountable.

See how Lowvyn protects your data | Read our full terms | Get the free browser extension

Yes, I Built Lowvyn With AI. Here Is What That Means.

Start tracking prices for free