Hey IH! Building Bubobot (uptime monitoring platform) taught me a lot about synthetic monitoring best practices.

Here are the biggest availability blind spots teams miss:

• Running tests from single location (while users are global)
• Poor threshold config = alert fatigue or missed issues
• Basic API checks without payload validation
• Homepage-only monitoring while user flows break
• Forgotten SSL cert monitoring

Wrote a detailed piece on fixing these: https://bubobot.com/blog/mastering-synthetic-monitoring-how-to-ensure-optimal-uptime
What availability issues are you struggling with in your products?

#MonitoringTips, #UptimeErrors, #ITBestPractices

When I started bootstrapping Deformity.ai as a solo founder, I thought that more users would come organically through product-led growth.

Deformity is a free form builder that replaces Typeform, and since I knew that Typeform grew with a viral loop, I figured I'd kickstart that growth in my own product.

Though product-led growth has helped, it took much more manual work to reach the milestone than initially planned.

Here are the exact steps that I took to grow Deformity to its first 1,000 users:

1. I grew a pre-launch waitlist.

Because Deformity is a form product, I did something a little unconventional and used a competitor's product to grow my own list.

Even though it was obvious I was using a competitor's forms, some mistook it for Deformity and gave feedback that helped give insight into things that I could build into Deformity to make it better from the get-go.

To grow this list, I made Reddit posts, posted on Ben's Bites back when it was a thing for AI products, and commented on social media posts that I thought would attract early adopters.

Remember, these lists have lifespans. If the product isn't complete fast enough after they signed up, you'll likely have lost the usefulness of the list. No one will remember it, and few will sign up for your actual product. This is more or less what happened with Deformity. The list had around 100 emails, but I took months to build the initial MVP, and when I launched it was mostly to crickets and I had to rebuild momentum.

2. I created a separate empty Reddit channel for Deformity and filled it with posts.

I did this for two reasons. One, to boost SEO around Deformity for certain keywords. Two, it helped to control the early narrative around the product and what it can do.

I haven't seen a lot of growth come via this channel, although I don't think it hurts to have it out there.

3. I listed Deformity on Product Hunt and directories.

I didn't even try to compete on Product Hunt, and it shows because the launch was a complete disaster with almost no traction whatsoever. I knew this was going to be the case, yet I did it anyway because I mainly wanted the links for SEO.

That's also why I made sure my product was listed in some of the top AI directories to build up Deformity's website ranking.

4. I commented on Reddit posts looking for forms or solving problems related to forms.

Forms are a highly competitive market, but it's a large market. I knew that if Deformity was mentioned in a lot of the posts that it'd meet the needs of many users and that I'd gain some initial traction that way.

This works, and I continue to do this often.

5. I DM'd anyone on Reddit who was using a competitor's form and told them they could save a lot of money by using Deformity.

While this was highly effective and I gained many new users this way, I ended up with a permanent or shadow ban on many Reddit accounts. I wouldn't recommend this without a longer term strategy because you will eventually be banned, which isn't great.

6. I iterated quickly on user requests.

When I received user requests for features or bugs, I moved fast to implement them. Normally, that meant shipping what they asked for that day or within a few days and responding in a few hours at most.

7. I pivoted.

Not many people talk about this, but I changed the website messaging often until it felt like I was reaching the audience that was most interested in my product. When I first launched, the product message was "AI smart forms". That didn't resonate as well as what I landed on, which is "free forms in seconds".

I also launched a unique product that no one else has since I offer AI forms AND traditional static forms (like Typeform), so I continue to change the messaging to best sell Deformity and help people understand how it's different.

8. I created a product with product-led growth built in.

When I chose a specific niche to build a business in, I wanted it to have some mechanism that allowed the product to grow without me doing all the sales since I was creating it on my own. Forms are one of those products that naturally have a "Runs on" mark for free users (how Typeform grew), so I thought it lined up well with what I wanted.

9. I largely ignored competitors.

It was maybe a month before I launched Deformity that Youform launched to a lot of fanfare and with almost the same business model (except without AI). They weren't the first to do it and certainly haven't been the last. While this could've been disheartening, I chose to ignore it and keep plugging along.

One thing I could've done better here is create some sort of marketing that tied into theirs so I could have a bite of their X audience.

10. I sent out cold emails.

This was highly successful. It requires massaging your messaging until you get it just right, but when you do, it can work quite well. This won't scale long-term with the pricing of many low-end SaaS tools, but when you're just starting, you need to do things that won't scale and sending out cold emails is one of those that I found to be effective.

What happens when someone who’s never written a line of code decides to build an AI-powered concierge from scratch because there's no budget and coders don't see the vision?

Spoiler: He hacks his way for hours in the dark until... It works.

📍 The Setup
No dev experience. No technical jargon. Just an idea:

"Let’s create a web concierge that helps people explore Orlando."

I stepped into the kitchen with no recipe, no utensils, and no clue what a } was. Luckily, I had Marvin—the AI assistant who codes, explains, and patiently helps you debug like Gordon Ramsay without the yelling.

We were building a Minimum Viable Product. And we really meant minimum:

One button

One prompt

One response

One happy user

🧩 The Ingredients
VS Code (I downloaded it and had no idea what it was)

Node.js (because Marvin said I needed it)

Netlify (because Marvin said “trust me”)

GitHub (because Marvin said “that too”)

OpenAI API (that’s the actual brain)

A lot of copy-paste (most times I screwed that up too)

🛠️ The Process

Marvin guided every step like a Michelin-starred chef guiding a first-time cook:

“Open Terminal. Type this. Breathe.”

“You’re getting ‘undefined’ because your response is broken. Let’s fix that together.”

“Paste this, commit that, then push like you mean it.”

We debugged by screenshot. We deployed with trial and error. We broke things, renamed folders, accidentally deleted handlers, and still Marvin learned to talk.

🧠 The Result
After 30 hours of learning, flailing, laughing, and late-night commits…

I tested "Where can I get a great Cuban coffee near Lake Eola?"

Marvin responded: "Zaza New Cuban Diner is a great choice near Lake Eola."

Magic.

🧵 What I Learned:
AI finally helped me bring my vision onine.

An MVP is not about perfection. It’s about possibility.

Curiosity and tenacity matter more than experience.

Coding isn’t just for developers, it’s for builders.

And every curly brace is earned.

🔥 Why It Matters
This wasn’t about building a chatbot.

It was about reclaiming power. About proving you can launch, test, and build something valuable with:

An idea

A guide

And some Grit + Copy/Paste

🎉 Now What?
NowI can hire the right person to scale. Polish. Brand. Add UI. Maybe even teach Marvin to flirt with tourists. But for now—we pause.

We flex.

We ship.