PrCheck

Last week I posted on several platforms about shipping code I did not really understand and asked if I had lost the plot. Dozens of you basically said "this is me," which was equal parts reassuring and alarming.

The thing that stuck with me from the replies: can you stop a landslide?. I had not thought about it that way, and it changed how I am framing the problem.

So I did the dumbest possible experiment. Before merging anything this week, I made myself re-read every line of code ;)

This is me thinking out loud, so bear with me.

A couple weeks ago I shipped a change I did not really understand. The AI wrote it, I skimmed it, tests passed, it merged. Then it broke, and opening the file I felt this out of body thing of not recognizing my own work. My name was on the commit. No memory of the decision.

I laughed it off. Then I noticed I do this constantly now.

Here is the shape of it. For years the slow part of building was writing code, so we threw everything at that. Autocomplete, Copilot, agents that write more in a minute than I read in ten. Writing is basically free now. But reading was always the real bottleneck, and nobody made that faster. So there is a growing pile of code that works, passes every test, and is understood by no one. People are calling it comprehension debt. It is technical debt that does not even look messy. You find out it is there when something breaks and the whole team reads the source together for the first time.

The reason I think there might be a business here: nobody measures it. We track how fast we ship a dozen ways. None of it measures whether anyone still understands what got shipped.

And here is where I talk myself out of it. Maybe it is not painful enough to pay for. Maybe it feels like surveillance the moment you productize it. Maybe the answer is just "prompt better." Maybe agents get so good nobody cares whether a human understands the code at all.

Then I open my editor, look at code I supposedly wrote yesterday, feel nothing, and think: no, this is real.

If you build with AI in the loop, does this itch you too, or have I lost the plot?

I didn’t start with the idea of building a startup.

I started with a pattern I kept seeing:
small EU medtech teams struggling with MDR Annex II, not because they don’t understand the regulation, but because the documentation process itself is fragmented, inconsistent, and stressful.

Annex II isn’t one document.
It’s a structured story spread across device descriptions, risk references, verification summaries, GSPR tables, and clinical evidence. In practice, that story often lives across folders, Word files, Excel sheets, consultant comments, and half-finished templates.

What I kept seeing:

• Teams staring at a blank Annex II template, unsure where to start

• Overdocumentation “just in case”

• Inconsistencies between sections that create notified body questions

• Knowledge stuck in people’s heads instead of in the structure

That’s the problem I’m trying to address.

What I’m building (and what I’m not)

I’m building a small web app that helps teams structure and draft MDR Annex II technical documentation step by step.

It is not:

• a QMS

• an automated compliance tool

• a replacement for regulatory responsibility or consultants

The goal is much simpler:
reduce blank-page stress, enforce structure, and improve consistency across Annex II sections.

Think of it as a guided workspace that helps you say the right things in the right place, while still requiring human judgment and accountability.

Why I think this matters

From what I’ve seen, notified body feedback is often not about missing tests or wrong standards, but about:

• unclear intended purpose wording

• weak linkage between risk, benefit, and performance

• inconsistent terminology

• unclear justification narratives

These are structural and narrative problems, not scientific ones.

If better structure alone can reduce back-and-forth with NBs, that’s already a win.

Where I am right now

This is very early.

• The concept is defined

• Annex II is mapped section by section

• I’m validating whether the pain is real enough to justify a dedicated tool

Before going further, I want input from people who actually work with MDR.

What I’d love feedback on

If you work in regulatory affairs, quality, or as a PRRC or consultant, I’d love to hear:

• Which Annex II sections cause you the most friction in practice?

• Where do you see teams consistently over- or under-document?

• What would you never trust a tool to help with?

• What would make you say “this might actually save time”?

No pitching, no links, just learning.

If this resonates, feel free to comment or DM.
I’ll share what I learn as I go.

A/B testing is supposed to make decisions easier—pick the variant with the best results and move on. But what happens when your ‘winning’ variant suddenly stops performing?

It’s a common issue. A test shows a clear winner, you implement it, and then conversion rates drop. Why? Maybe user behavior changed. Maybe external factors (seasonality, traffic sources) influenced the test. Or maybe the test didn’t run long enough to reach real statistical significance.

A/B testing isn’t a one-time fix; it’s a continuous process. That’s why SplitSense doesn’t just declare a winner—it keeps testing, adapting, and refining. Because what works today might not work tomorrow.

For the past few weeks, every day brought a little rush—another email notification, another name on the waiting list. It felt like momentum was building, like SplitSense was finding its early believers. Then today… nothing. No new subscribers. Just silence.

It’s a strange feeling as a solo founder. One day, things are moving, and you feel unstoppable. The next, it’s as if all the progress has hit an invisible wall. But maybe this is where the real work begins—not in celebrating small wins but in pushing forward when the excitement isn’t automatic. So, what’s next? More outreach, better messaging, and refining the product. Because the only way to break the silence is to keep showing up.

Building a product is exciting but also intimidating. With SplitSense, I decided early on to share the journey publicly—every win, every struggle, and even the missteps. Why? Because transparency builds connection. Especially on days like these, where I'm stuck in a piece of code and can't figure out how to solve it.

It hasn’t been easy. Writing about challenges, like how to define the MVP or code struggles, makes me feel exposed. But the feedback and encouragement I’ve received have been invaluable. It’s helped me shape the product with real input from people who understand the problem I’m solving.

Building in public keeps me accountable and reminds me that startups are about people, not just products. If you’ve ever thought about doing the same, I can’t recommend it enough. And if you want to follow SplitSense’s journey or join the waiting list, visit subscribepage.io/splitsense. Let’s build this together.

//EDIT 27/1: And already posting my update here lead to a great improvement. The FAQ is no longer a Lorum Ipsum :)

When you’re building something new, it’s easy to get distracted. There’s always a detail to tweak, a feature to add, or a “what if” that pulls your attention in a hundred different directions. With SplitSense, I’ve learned just how tempting it is to get lost in the weeds—but I’ve also realized the importance of staying focused on what truly matters.

SplitSense is about simplifying A/B testing, and yet, I’ve found myself obsessing over things like button colors and fonts instead of tackling the core functionality. It’s a trap that’s easy to fall into: spending time on things that feel productive but don’t actually move the needle. When I catch myself heading down a rabbit hole, I step back and ask, “Does this make the product better right now?” More often than not, the answer is no.

Focusing on the essentials has been a game-changer. By prioritizing what truly matters—automating A/B testing and delivering actionable insights—I’ve been able to make real progress. It’s not about getting everything perfect; it’s about building something valuable and getting it into users’ hands as quickly as possible.

The SplitSense waiting list is live, and we’re getting closer to launching the beta-demo. If you want to see what we’ve been working on, join the journey at subscribepage.io/splitsense. I can’t wait to hear your feedback as we bring this vision to life.

Building a startup is never a straight line—it’s a rollercoaster. While I'm excited about the early interest in SplitSense, THANKS TO ALL WHO SUBSCRIBED TO THE WAITLIST ALREADY!!, I want to share some behind-the-scenes struggles in getting the product ready and defining what the MVP should actually look like.

The Struggle Is Real

When I first started writing the code for SplitSense, I underestimated just how complex A/B testing could get. Sure, the concept seems simple—show two versions of a page and pick the one that performs better. But when you start diving into the nitty-gritty, there are a million questions:

• How do we define "significance" in the test results?

• How do we handle edge cases, like when there isn’t enough traffic to reach meaningful insights?

• What’s the best way to present actionable data without overwhelming users?

As a solo founder (at least for now), balancing coding, testing, dreaming and planning is tough. Some days, it feels like you’re stuck in a loop of “What if this doesn’t work?” But here’s the thing: every struggle feels worth it when you see progress.

Less Is More?

One of the hardest decisions in this process was deciding what features to focus on for the MVP. It’s tempting to include everything and try to deliver a fully polished product from day one, but that’s not realistic. Instead, I had to ask myself:

• What’s the core problem we’re solving?

• What’s the simplest version of the solution that provides real value?

For SplitSense, that meant focusing on these key features:

1. AI-Driven Variant Creation: Automatically generate A/B test variants for landing pages.

2. Automated Testing: Run tests with minimal user input.

3. Actionable Insights Dashboard: Provide clear recommendations based on test results.

Anything beyond that—like integrations with major CMS platforms, custom analytics, or advanced segmentation—can come later. Right?

Lessons Learned So Far

1. Perfection Is the Enemy of Progress: I’ve learned that it’s better to launch with a smaller feature set and improve over time than to wait for everything to be perfect.

2. User Feedback Is Gold: The subscribers who’ve already signed up for early access are shaping the direction of the platform. Listening to them will help us build something that truly solves their problems.

3. Keep It Simple: A/B testing can be overwhelming. My goal is to make it so intuitive that anyone can use it without a manual.

What’s Next?

The next step is to finalize the beta-demo and share it with the early subscribers. I’m thrilled (and a bit nervous) to see how SplitSense performs in the real world.

To everyone who has signed up already: thank you! Your support means everything. If you haven’t joined the waiting list yet, now’s the time: https://subscribepage.io/splitsense.

Final Thoughts

Building SplitSense is a journey, and every challenge is an opportunity to make the product better. I’ll keep sharing updates as we get closer to launch—both the wins and the lessons learned along the way.

Let me know: What’s one feature you’d love to see in an A/B testing tool?

The wait is over (well, almost). I'm thrilled to announce that the waiting list for SplitSense is now live! If you're as excited about revolutionizing how businesses optimize their landing pages as I am, then this is your moment to jump on board.

Why SplitSense?

For too long, A/B testing has been complicated, time-consuming, and often out of reach for small and medium-sized businesses. SplitSense changes the game.
Our platform leverages artificial intelligence to:

• Instantly generate optimized variants of your landing pages.

• Automate testing processes with ease.

• Deliver actionable insights to help you make smarter decisions, faster.

Imagine increasing your conversion rates without spending hours analyzing data or hiring expensive consultants. That’s what SplitSense is here to do—empower businesses to achieve better results effortlessly.

The Beta Is Coming Soon

We’re just weeks away from opening the beta-demo to a select group of users. This is your chance to be part of our early community and experience the power of SplitSense firsthand. By signing up for the waiting list, you'll:

• Get exclusive early access to the beta-demo.

• Be among the first to shape the future of SplitSense with your feedback.

• Gain insights into how AI-driven A/B testing can supercharge your business.

Who Should Sign Up?

SplitSense is perfect for:

• Digital marketers looking to maximize ROI.

• Small business owners who want results without technical headaches.

• E-commerce entrepreneurs striving for higher sales.

Join The Journey

I'm building SplitSense with the vision of making advanced A/B testing accessible to everyone. But I can’t do it alone—I need YOU! Your feedback will help me fine-tune the platform and create a tool that truly meets your needs.

👉 Sign up now at https://subscribepage.io/splitsense