That first $10K MRR usually come from a series of small, imperfect steps. Tweaking your offer after a few calls, realizing what people actually care about, adjusting as you go, and sometimes starting over more than you expected. It’s a mix of doubt, small wins, and figuring things out in real time. This month at Gildre, we’re focusing on those in-between moments—the ones that don’t get posted but make all the difference. Sharing real experiences, so you can take what’s useful and apply it to your own build.

Jon Yongfook didn't have a massive marketing budget or a viral loop. He hit $10k MRR in roughly one year by acting as a "distribution machine" and using "Engineering as Marketing" as his primary weapon.

1. The "Anti-MVP" Shift: Solving a Specific Workflow

Bannerbear is an API that generates social media images and videos. Before it was a success, Jon launched several "viable" products that failed because they were too broad.

• The Tactical Pivot: He realized marketers were wasting hours manually typing text into Canva to make 50 different variations of an ad. He stopped building a "design tool" and built an automated workflow (an API).

• The Result: He moved from selling a "feature" to selling "time saved on Tuesday afternoons."

2. Engineering as Marketing (The "Sidecar" Strategy)

This was Jon’s secret weapon. To get traffic without waiting 6 months for SEO blog posts to rank, he built mini-tools that solved tiny problems for his target audience (developers and marketers).

• The Tool: He created a "Smart Crop Demo" and a "Face Detect AI Demo."

• Why it worked: Developers searching for "how to auto-crop an image" found his free tool. Once they saw how well it worked, the "Bridge" to his paid API was obvious. It provided immediate value and ranked on Google almost instantly.

3. Aggressive Pricing (The $9 Trap)

Jon is vocal about the fact that he initially underpriced his product.

• The Tactical Shift: He realized that at a low price point, he was attracting users who required too much support. He raised his prices to target agencies and mid-tier companies who had higher budgets and lower churn.

• The Result: This allowed him to reach $10k MRR with a much smaller, more manageable customer base.

4. The "Unscalable" Acquisition (Founder-Led Distribution)

While building, Jon didn't hide. He used Transparent Building on Twitter and Indie Hackers.

• The Tactic: He shared his exact revenue numbers, his coding struggles, and his marketing experiments.

• Why it worked: This created "hand-to-hand" trust. People weren't just buying an API; they were supporting a founder they knew. This early community acted as a zero-churn foundation that carried him to $10k MRR.

The Summary of Bannerbear’s $10k Leap

If you’re thinking about Product-Led Growth, this roundtable on Achieving Product-Market Fit is honestly a really important watch. It breaks down the real, messy process of building something people actually want — starting with the problem (not the solution), testing MVPs early, getting honest feedback from beta users, and paying close attention to what users actually do after launch.

It’s a good reminder that PLG isn’t about hype — it’s about learning fast, measuring what matters (retention, engagement, those “aha” moments), and being willing to pivot when the data tells you to. If you want your product to drive growth instead of constantly pushing from the outside, this conversation hits the fundamentals in a very practical way.

Founders’ Toolkit - Practical Considerations Before Choosing an AI Tool

By: Matt Savare and Bryan Sterba, Partners at Lowenstein Sandler

Note: This article was meant to replace the one received in the initial newsletter delivered on April 6th. We have since updated it on the web version to reflect the timing.

AI tools can feel like a superpower for early-stage startups. They write copy, analyze volumes of data, generate code, and quickly and efficiently automate work that used to take weeks. And they can help startups punch much higher than their weight class. But choosing an AI tool is not merely a product and technological decision. It is also a legal, security, and strategic decision that can shape your company’s future. Get it wrong, and you risk exposing sensitive data, compromising your intellectual property (IP), and creating legal liability before you ever find your footing.
Here are some practical considerations you should evaluate before adopting any AI tool.

1. Should You Use an Enterprise or Public AI Tool?

The first question you must consider is whether to use an enterprise or free AI tool. Compared to free, public AI tools, enterprise tools typically provide enhanced data security, stronger confidentiality protections, more favorable IP rights, and most importantly, substantive recourse against the AI tool’s provider when material issues arise. On the down side, they cost money, which may be an issue while bootstrapping.

To evaluate this issue, you need to decide the purposes for which you will use the AI tool and what information your team will input into the AI. Early-stage teams often indiscriminately paste customer data, product roadmaps, source code, fundraising materials, employee information, internal strategy
documents, and other highly-sensitive materials into the AI. If so, you should use an enterprise version.
Ultimately, the cost of an enterprise license is a fraction of the cost of a single data breach, regulatory fine, or lost customer relationship.

2. Is the Data You Input into the AI Tool Confidential?

Most free, publicly-available AI tools reserve broad rights to store, review, and use your inputs to train and improve their models. Although this may not be an issue if the data you input into the AI tool is aggregated and anonymized, chances are that there is someone on your team who will input something sensitive into the AI. This can create a serious problem in a number of respects. First, information that you thought was confidential may no longer be treated that way legally. Second, if any of the information
is personal data, data concerning children, or other form of regulated data (e.g., health or financial information), you may be violating one or more laws. Third, if any of the information is covered by a confidentiality agreement you have with a third party, you may be breaching that agreement by inputting into an AI tool that does not adequately safeguard the information.

Therefore, it is critical that you review the agreement, terms of use, terms of service, or other document governing your use of the AI tool. If you anticipate inputting any sensitive, confidential, or regulated information into the AI tool, you must ensure that the agreement contains a confidentiality clause precluding the AI vendor from disclosing or using your data for any purpose other than providing you access to, and usage of, the AI tool.

If the AI tool will serve a mission-critical function within your organization, you almost certainly need to use an enterprise version, diligence the AI vendor, and negotiate appropriate data security, confidentiality, and IP provisions.

3. Who Owns the Output and is it Protectable?

Most AI tools (even public AI tools) stipulate that you own the output generated from your prompts.
However, keep in mind that even if you own the outputs, the AI vendor may still be permitted to train on your prompts and inputs, or generate similar outputs for competitors. If you are using the AI tool to generate important IP that gives you a competitive advantage, this may be unacceptable to you. Again, be sure to confirm what the governing agreement with the AI vendor states and negotiate accordingly.
A common misconception of early-stage founders is that ownership automatically means that the outputs are protected by some form of IP. This is not always true and depends on the type of IP, the level of human involvement, and the jurisdiction. Under U.S. law, the following rules apply:

• Copyrights – if an AI tool generates the content and a human did not meaningfully shape the final product, there is no copyright.

• Patents – an AI tool cannot be an inventor, and a human must be the inventor even if the AI assisted.

• Trade Secret – AI-generated output can be protected as a trade secret provided the output derives economic value from not being public and it is subject to reasonable efforts to keep it secret. So,
  there is heightened risk if you use public AI tools or allow vendors to reuse your inputs or outputs.

• Trademarks – AI-created branding (e.g., logos, names, and slogans) can be protected as trademarks as long as they are distinctive, you use them in commerce, and no one is using a mark that is confusingly similar.

The Bottom Line

AI tools can dramatically accelerate early-stage startups, but only if used thoughtfully. The right tools, used the right way, are worth the initial investment. You should conduct diligence on your AI vendors and review any agreements before signing or clicking through. This can prevent expensive surprises later, especially as your company grows, raises capital, or gets acquired.

Lowenstein Sandler LLP is a national law firm with over 400 lawyers based in New York, Palo Alto, Roseland, Salt Lake City, San Francisco, and Washington, D.C. We represent clients in virtually every sector of the global economy, with particular strength in the areas of technology, life sciences, and investment funds. Authors Matt Savare and Bryan Sterba advise founders, growing companies, and established global corporations on the deals, agreements, and strategic decisions that define their trajectory. From commercial contracts and IP transactions to AI and emerging technology, Matt and Bryan bring a practical, business-first approach to startup legal work.

If you have any questions for Matt or Bryan you can reach them at:

Matt Savare Bio
Bryan Sterba Bio

Matt Savare LinkedIn
Bryan Sterba LinkedIn

When it comes to your first $10K MRR whether you’re bootstrapped or venture-backed, it’s easy to focus purely on growth, revenue, or what’s working to immediately move the needle. But building something real also means making better decisions as you go - knowing when to move fast, when to iterate, and when to bring in the right people and tools to support you. The founders who get there aren’t just experimenting on the surface - they’re building with intention underneath it all.

And that’s exactly what we’re building at Gildre: a space where you don’t just chase milestones, but understand how to sustain them alongside other founders & operators. If you’re in that process of your growth, you can book a conversation to learn more with Managing Partner, Taiga Gamell, here.

Cheers to the month ahead,
Eliana