Category: Entrepreneurship

Founder journey, startup lessons, and business strategy

What You’re Really Avoiding Isn’t the Work

Everyone has a version of this. A category of work that sits on the to-do list for weeks, then months, slowly accumulating guilt. For some founders it’s legal. For others it’s HR, compliance, or investor reporting. For me, it’s always been accounting.

Not because I can’t do math. Because every time I opened QuickBooks, I’d feel the weight of everything I didn’t understand, and I’d close the tab. There’s always something more urgent than confronting what you don’t know.

This week I finally sat down and did all of it. Reverse-engineered spreadsheets. Audited our QuickBooks accounts. Found missing payables. Fixed miscategorized transactions. Worked through international currency adjustments. Even handled an off-the-books equity correction I’d been dreading for longer than I’d like to admit.

And here’s the part I didn’t expect: it was actually kind of fun.

The difference wasn’t discipline. It was having AI as a collaborator. And the reason that mattered has nothing to do with accounting specifically.

The real barrier is shame

Think about the task you’ve been avoiding. Now think about why.

It’s probably not because the task itself is impossibly hard. It’s because there’s a gap between what you know and what you’d need to know to do it confidently, and closing that gap feels expensive. You’d have to ask someone. That someone is busy, or expensive, or both. And the questions you need to ask feel like they should be obvious.

That was my relationship with accounting for years. Accountants always seem busy. When I’d get on a call with mine, I’d feel the clock ticking. Every question felt like it should be obvious. Do I really need to ask what a trial balance is? Can I admit I don’t understand why this line item is negative? Is it okay to not know the difference between cash-basis and accrual?

So you nod along, say “makes sense,” and leave the call having learned nothing. Then you avoid the whole topic for another month.

This is the shame barrier. It’s not a knowledge problem. It’s a help-access problem. The help exists, but the social cost of accessing it is high enough that you just… don’t.

What happens when the shame disappears

When I sat down with Claude Code this week and started working through our financials, I could ask anything. Literally anything.

“What does this column mean?” No judgment. “Why is this number negative when we received money?” Clear explanation. “Walk me through how this journal entry should work.” Step by step, as many times as I needed.

I went deep on things I’d been skating past for years. The nuances of our P&L statement. How the balance sheet connects to the trial balance. Why certain transactions were showing up in the wrong categories. What our cash flow statement was actually telling me versus what I assumed it was telling me.

Each question led to a better question. And because I wasn’t worried about wasting someone’s time or looking dumb, I kept going. I’d ask a follow-up, then another, then branch into something related. It was the first time accounting felt like learning instead of an exam I was failing.

If you’ve ever had a mentor who made you feel safe asking the dumb questions, you know how much faster you learn in that environment. AI gives you that dynamic on demand, in any domain, at any hour.

The concrete results

This wasn’t a vague learning exercise. I worked through real problems in our actual books:

Reverse-engineered inherited spreadsheets. We had several financial spreadsheets maintained by different people over time. I fed them to Claude and asked it to explain what each one was tracking, how the formulas worked, and where there were inconsistencies. It found things that had been wrong for months. If you’ve ever inherited a spreadsheet from someone who left the company and spent hours trying to figure out what it was supposed to do, AI turns that from hours to minutes.

Audited QuickBooks categories. Transactions miscategorized across multiple accounts. Expenses in the wrong cost centers. Payables missing entirely. Claude walked me through each one, explained what the correct category should be and why, and helped me make the corrections.

Handled the stuff I’d been avoiding. International currency adjustments. An equity correction I didn’t fully understand the accounting treatment for. Reconciliation of accounts that hadn’t been reconciled in too long. These are the kinds of things where I’d normally email the accountant, wait three days, get an answer I half-understood, and still feel uncertain about whether it was done right.

Thought through the strategic questions. Beyond the bookkeeping, I used the conversation to think through bigger questions. I’ve thought about managing cash flow before, but this was different. What are our actual options right now? What interest rate is expensive versus reasonable for our situation? What are the trade-offs between different funding approaches? These aren’t strictly accounting questions, but they live in the same “financial stuff I’m uncomfortable with” bucket, and having a patient conversation partner made them approachable.

The pattern worth noticing

Here’s what I want you to take from this. It’s not “use AI for accounting,” although you should.

Every business owner has domains they understand well and domains where they’re faking it. For me, the product development, marketing, and technical infrastructure are comfortable territory. Finance has always been the thing I know I should understand better but never prioritize learning. It’s a version of the fear of the unfamiliar that I think most founders carry around quietly.

AI doesn’t replace the expert. I still need a CPA for tax strategy and compliance. But it fills the gap between “I know nothing” and “I know enough to have a productive conversation with my accountant.” That middle layer of competence is what most people skip, and it’s exactly where AI excels.

Before this week, my accounting approach was “send everything to the accountant and hope for the best.” Now I actually understand what’s in our books. I can read a P&L and know what I’m looking at. I can spot when something looks wrong. That upgrade happened because the learning barrier dropped to zero.

Apply this to your thing

This keeps happening. Tasks I’ve been dreading turn out to be approachable, even enjoyable, once I have a collaborator that’s patient, knowledgeable, and available whenever I’m ready to work. It happened with growth engineering. It happened with the small automations that add up. Now it’s happened with accounting.

The common thread is that the barrier was never ability. It was the friction of getting help. AI removes that friction, and suddenly the things you’ve been avoiding become the things you’re making progress on.

So here’s my challenge to you: think about the task that’s been sitting on your list the longest. The one you keep bumping to next week. Ask yourself whether the problem is really that the task is hard, or whether the problem is that you don’t have a safe, low-cost way to close your knowledge gap.

If it’s the second one, you might be surprised at what happens when you just start asking questions.

February 13, 2026
Building a Personal Knowledge Base: How I Created a Semantic Search Engine Over Everything I’ve Ever Made
I’ve been creating content for years. YouTube videos, blog posts, tweets, podcast appearances, internal docs for my company. Thousands of pieces scattered across platforms and folders.

Here’s the problem: I can’t remember what I’ve said.

Not in a concerning way. In a “did I already share that framework?” or “what was that thing I said about distribution vs product?” way. My past content exists, but I can’t access it when I need it. When I sit down to write something new, I’m starting from scratch instead of building on foundations I’ve already laid.

The Inspiration

I was listening to a podcast where Caleb Ralston (a personal branding creator on YouTube) mentioned that his team had built an “AI database” of all his historical content. They transcribed every video he’d ever appeared in and turned it into something searchable. It let them understand his existing talking points, find frameworks he’d already developed, and maintain consistency across content.

The concept stuck with me. What would it look like to build something similar for myself?

What I Built

A local semantic search engine that can answer questions about my own content. The entire system runs on my laptop. No cloud services, no API costs after setup, complete privacy.

The stack is surprisingly simple:
- ChromaDB for vector storage
- Ollama for local embeddings (nomic-embed-text model)
- Python script to ingest and query
- Markdown as the universal format
Total setup: maybe 200 lines of code.

How It Works
1. Collect content – YouTube transcripts (downloaded via yt-dlp), blog posts, docs, anything in text form
2. Chunk it – Split documents into ~500 word segments with overlap
3. Embed it – Convert each chunk to a vector using Ollama locally
4. Store it – ChromaDB persists everything to disk
5. Query it – Semantic search returns relevant chunks for any question
```
# Ingest all content
uv run build-kb.py --ingest

# Ask questions
uv run build-kb.py --query "What have I said about content systems?"
uv run build-kb.py --query "My thoughts on distribution vs product"
```
The “semantic” part matters. I’m not doing keyword matching. When I ask about “content systems,” it returns chunks that discuss workflows, automation, and publishing pipelines—even if those exact words aren’t used. The embedding model understands meaning, not just strings.

The Obsidian Connection

Here’s where it gets interesting.

My entire working directory is a folder of markdown files. Blog posts, notes, drafts, transcripts—all .md files in a structured hierarchy. That folder is also an Obsidian vault.

Obsidian gives me:
- Visual browsing – Navigate content through a nice UI
- Linking – Connect related ideas with [[wiki-style links]]
- Graph view – See how concepts cluster together
- Search – Quick full-text search when I know what I’m looking for
The knowledge base adds:
- Semantic search – Find content by meaning, not keywords
- Cross-reference discovery – “What else have I said that’s similar to this?”
- Topic clustering – Analyze patterns in what I talk about most
They complement each other. Obsidian for browsing and organizing. The knowledge base for querying and discovering.

What I Discovered

After ingesting ~400 chunks from my content, I ran an analysis to find topic clusters. The results were illuminating:

Topic Frequency
Claude Code / AI automation 86 mentions
Content systems & workflows 75 mentions
Marketing & business 106 mentions
Founder productivity / goals 62 mentions

The phrase “claude code” appeared 38 times in my personal brand content. “Content” appeared 131 times. These are the themes I return to constantly.

More useful than the raw counts were the semantic clusters. When I queried “What have I said about content systems?”, I got back chunks from:
- A blog post about growth engineering with Claude Code
- A YouTube video called “Creating a Content System”
- Internal documentation about creative direction
Content I’d forgotten I made. Ideas I’d already articulated that I can now build on instead of recreating.

The Broader Pattern

This is part of something I’ve been calling “growth engineering”—treating marketing infrastructure like software infrastructure. The knowledge base is one component.

The full system looks like this:
```
Working Directory (Obsidian Vault)
├── posts/           # Blog content
├── content/         # Thought leadership drafts
├── knowledge-base/  # Vector DB + scripts
│   ├── youtube-transcripts/
│   ├── chroma-db/
│   └── build-kb.py
└── products/        # Product pages and docs
```
Everything is markdown. Everything is version controlled. Everything is queryable.

When I want to write something new:
1. Query the knowledge base: “What have I said about [topic]?”
2. Review existing content in Obsidian
3. Build on what exists instead of starting fresh
4. Publish through the same markdown → WordPress pipeline
The AI isn’t writing my content. It’s helping me remember and organize what I’ve already created. The knowledge base becomes institutional memory for a one-person operation.

How to Build Your Own

If you want to try this, here’s the minimal setup:

1. Install Ollama
```
brew install ollama
ollama serve
ollama pull nomic-embed-text
```
2. Create the ingestion script

The core is maybe 100 lines. Collect documents, chunk them, embed them, store them in ChromaDB. The full script is in my knowledge-base repo.

3. Point it at your content

YouTube transcripts are easy:
```
yt-dlp --write-auto-sub --sub-lang en --skip-download \
  "https://www.youtube.com/@your-channel"
```
Markdown files just need to be in a folder. The script recursively finds them.

4. Query away
```
uv run build-kb.py --query "your question here" -n 10
```
The embedding model runs locally. No API keys needed after you pull the model. Completely private—your content never leaves your machine.

The Meta Layer

There’s something recursive about using AI to build the system that helps me leverage AI.

Claude Code helped me write the ingestion script. It helped me debug the VTT parsing for YouTube transcripts. It helped me analyze the topic clusters. Now the knowledge base feeds context back into Claude Code when I’m working on new content.

The tools build the tools that improve the tools.

That’s the pattern I keep returning to. Not “AI writes my content” but “AI amplifies my ability to create and connect my own content.” The knowledge base doesn’t have opinions. It has receipts—everything I’ve said, searchable by meaning.

For someone building a personal brand, that’s the foundation. Know what you’ve said. Build on it. Be consistent AND repetitive but in unique ways. Let the system remember so you can focus on what’s new.
January 27, 2026
Merchant Cash Advance to Annual Interest Calculator

in eCommerce, Merchant Cash Advances are common financial offereings. Both Shopify and Amazon have integrated financial offerings based on your sales data. These offers are often quite appealing: borrow $10,000 and payback $11,000 in 6 months.

At face value these offers seem like a fair market rate like 10% in that case. However, it’s a bit deceiving.

If you had instead borrowed with a term loan, then as you pay down the principal of the debt the interest payments come down as well.

So, borrowing $10,000 at 10% APY, with monthly payments you’d pay back about $10,530. Not $11,000. That’s a big difference!

Here’s the calculator to find the equivalent APY rate for a merchant cash advance. Use this to more fairly compare the cost of credit and wether or not you’d be better off with credit cards

MCA to Annual Interest Calculator (Monthly Payments)

Enter your merchant cash advance details below:

Advance Amount:

Total Repayment Amount:

Term (in months):

February 17, 2025
Founder Fuel: How do you Manage your Business Finances and Cash Flow?
Sometimes a business is in growth mode and it makes sense to spend money to buy revenue, other times cashflow is king and the focus is on accounts receivable and payable. This stuff is hard and involves a bunch of things I rarely hear people talk about.

If you have your fingers into the accounting system like I do there are some tools to help make the bookkeeping just a little less tedious and give you visibility on where the money is coming and going.

Here’s the current software in my finance stack:
1. Dext – for automatically parsing invoices and receipts, it categorizes and sends into Quickbooks
2. Quickbooks Online – love it or hate it, it’s a defacto standard that’s hard to avoid for profit and loss, and balance sheets. and is the double entry accounting final resting place for all numbers.
3. A2X – tears open the payouts sent from Amazon and Shopify into the more granular transactions that make it up. These import into QBO where it’s then possible to see revenue and payment processing fees, returns and other transactions split out.
4. custom software – theres a few things that A2X can’t handle and for that we have some custom tools (nice to have a software dev in-house). It can help get COGS numbers across all sales channels, and display nicer reports.
5. Google Sheets – the most versatile tool for adhoc models, and tracking things. Flexibility and shareability is hard to beat.
6. Notion – notion is the home of all documented SOPs. These how to guides are incredibly helpful for making sure that recurring tasks are understood and done consistently each time.
7. Banks with virtual cards – virtual cards make it easier to give employees a card for paying expenses, and lower risk that if one card is compromised, a bunch of things will need to be updated. I’m using Relay in the US and Vault in Canada.
Is there anything I wish could be done better? Yes. The custom software is kind of a pain, mostly because so many systems it interacts with don’t have APIs and have to use fragile ways to get data.

And that’s it for this Daily Founder Fuel Journal entry.
July 21, 2024
Founder Fuel: How do you measure the success of your marketing efforts?

At different times in a business success means different things. Sometimes it’s measured in Likes and views, other times in ROAS or ACOS, other times it’s in brand recall. But as with most things in business, if you don’t measure you aren’t in control of it. So KPIs for your marketing efforts is an important aspect of understanding if things are working as intended.

I have been drawn to sales and direct marketing approaches in the past because the metrics are easy to collect. I think most small companies should start here. The direct testing enables quicker learning which can then form the backbone of a brand. Where as jumping straight into brand marketing usually requires a lot of assumptions.

With one-on-one sales you get immediate and clear annecdotal feedback about what customers want. Then expand with direct marketing ads to scale those up to 10,000+ people to further refine the messaging and targeting, and test the market further. Finally layering in the branding in a way that applies everything you’ve learned so far.

Easier said than done. People feel compelled to start with a logo.

As you get more sophisticated, something I have not yet explored is Marketing Mix Modeling approaches to measure the effectiveness of marketing. With a complex set of marketing channels in which to budget marketing spend this is probably the best way to assess how it all is affecting sales.

That’s how I think about measuring the success of marketing. This journaling prompt came from Daily Founder Fuel a very short daily newsletter that contains a journaling prompt for founders, entreprenurs and business owners.

July 10, 2024
Founder Fuel: How do you motivate and inspire your team?

Leaders can lead in different ways, but some of the best leaders have been truly inspirational. They may inspire people to invent new things, or produce better work, or just provide enough motivation to put in 5 extra minutes. The Big Hairy Audacious Goals (BHAGs) can work, but can also backfire if people don’t believe they’re possible.

It’s been really enlightening to see that clear vision and inspiration can break through walls, but that once those walls are broken others quickly follow behind. This happens in sport – the 4 minute mile – once broken more and more people started to do what was previously thought impossible. It also happens all the time in technology – Landing rockets, now that SpaceX has done it, the Chinese are copying it.

Inspiration is required to get the first people to unlock these barriers.

How do I inspire? I hope to inspire by setting a good example. By working hard. By being informed enough to provide good reasons for decisions. But it is hard. Inspiring is not a skill learned in high school, though perhaps inspirational writing should be taught right after persuasive writing.

I believe that more times than not inspiration comes from good story telling. The story you tell yourself or that you tell others.

It’s not something that I feel confident I have excelled at in practice. But through writing down these thoughts I have a little bit more clarity on how I might learn and improve.

For more inspirational journal prompts subscribe to Daily Founder Fuel

July 9, 2024
Founder Fuel: What new markets or customer segments could you explore?

The best markets and segments to explore are the ones that are adjacent with what you already have. The further you go the more you have to start from the beginning to find the audience and the less overlap you have with your existing marketing assets.

Much more preferable to push into new markets by expanding at the borders.

This land and expand type of strategy could be geographic, it could be demographic or it could be interest based.

Now, the thought of how to be deliberate about this comes to mind. How do you align people and keep the focus on markets and customer segments that can be efficiently won? Tie bonuses to winning certain types of customers over others? Create visualizations of the data?

July 8, 2024
Founder Fuel: What inspired you to start your business?

Inspiration comes from many places. The original inspiration for Psychedelic Water came from a desire to go after what a trend towards the opening up of psychedelics – how can we catch the wave with something in the comercial space. Delving deeper into the research revealed more reasons why there was a need for this product to be on the market.

Other ideas come from trying to solve problems. How can I solve for cheaper housing? How can I design a more interesting e-bike? How do I apply AI to improve productivity?

My current insipiration has gotten me to think about a blend of a holding company concept as a way to test out all the backlog of ideas I have. There’s so many things I want to be able to accomplish but not enough time or capital to. So I’m attempting to test the ideas out cheaply and quickly with a build in public approach and should anything prove useful, it can develop further under the umbrella of my company.

I think at a more fundamental level, I like to start businesses because I’m attracted to uncertainty. Life is more interesting when you are dealing with a level of not knowing everything. There’s always more to learn, always new situations and people to interact with. Ideas are everywhere and it’s just too compelling to not explore them.

July 7, 2024
Starting Daily Founder Fuel

This past week, I had an idea for an app. This idea came from an impulse to jot down some thoughts about my business challenges and how they needed to be written out, journaled, thought through, and developed further. I wanted to incorporate this into a daily practice, recognizing the value of writing. Everyone knows that through writing, ideas become more concrete, real, and memorable, as well as easier to share. So, writing was on my mind as I considered how to approach this.

I also wanted to maintain a balanced approach to my thought process. Some days are for strategic thinking, others for sales processes, numbers, finance, long-term growth, or professional development. As a founder or entrepreneur, it’s crucial not to fall into old patterns of focusing only on preferred areas but to address all necessary aspects that might otherwise be neglected. Having a structured approach ensures a balanced distribution of thoughts and developing ideas, preventing a single-minded focus and fostering a holistic view of the business.

I began searching for journaling apps tailored to entrepreneurs, addressing their specific concerns and questions to improve their business and life. Most journals available are generic, catering to a wide audience with personal goals and life thoughts. I wanted something more specific to business ideas. While I enjoy writing on paper, a physical book can be easily forgotten. To counter this, I decided to create an email newsletter that would appear daily, ensuring it remains visible and part of my routine. This way, it consistently prompts daily reflection without being easily hidden or forgotten.

After collecting journaling prompt ideas and quotes, I realized many turned into homework-like tasks, which, while interesting and fun, also served as valuable exercises. Questions about handling team conflicts, delegation, sales tactics, personal skill development, team motivation, and defining unique sales propositions kept me engaged. These prompts helped test my clarity of thought and understanding of various business aspects, ensuring I stayed sharp and well-rounded in my approach.

Seeing a need for such a resource, I launched a website, dailyfounderfuel.com, and created a newsletter signup so everyone could try it. I populated it with prompts scheduled out for the next several months. This experiment required minimal effort and low cost—around $40 initially and $10 monthly for email service, domain, and hosting. This small investment sets up an ongoing experiment to see if there’s interest in such a resource. If successful, I’ll have a valuable list of entrepreneurs and founders. If not, it’s a learning exercise. Either way, it’s a worthwhile endeavor. If this sounds interesting, check out dailyfounderfuel.com and sign up.

June 27, 2024
My Early AI Business Failure

The year was 2011, I was deep into affiliate marketing and AdSense with content websites. On the side I had created and launched a dozen different WordPress blogs.

Things were getting unwieldy. The backups, updates, monitoring, designing, and writing content for all these websites was a lot of work, and it easy to miss things that broke.

I knew that keeping the content fresh and updated was key to ranking on Google. Even back in 2011 I been already been blogging for over a decade.

I wanted to scratch my own itch and build some tools that could automate the management of these WordPress installations. As with all things that get automated I wondered about how big of a scale this could get. Would it be possible to manage 100 blogs? 1000?

At a certain scale, writing blog posts for these websites becomes an impossibility, so I started looking into an approach called content spinning. This used an earlier approach of AI techniques called Natural Language Processing to re-word and re-arrange other written content so that it appeared unique in the eyes of Google.

I built it and it worked!

This platform could crawl the internet and compile the latest news, and interesting data into dozens of fresh blog posts every day.

This was the first SaaS business that I built and launched with paying customers. I was pumped.

A customer could simply load in a domain name, a list of keywords to target for the blog content and select one of several available themes. The system would then install WordPress, configure it with users, themes and plugins and it would start the processes to crawl the internet for related content that could be re-purposed.

It was almost entirely hands off.

I used it to launch and run over 60 websites.

The launch went well, and I had a perfect number of users to build off of.

And then it happened.

The same month that I launched Automatic Blog Machine, Google rolled out a major overhaul of it’s search engine and it was insanely good at finding websites with content like what was possible to generate with the NLP approaches I was using. It was immediately able to flag and de-rank all these websites.

With no other better approaches for automated content generation, and better AI still a decade away the users slowly churned. The defeat was de-moralizing, and instead of pivoting this into what could have become a decent WordPress management and hosting service I lost the motivation to keep it going.

At the time I learned the wrong lessons from this experience:

Google can stomp you out of business in an instance

Timing is just part of the random chance involved – I lost.

However, with more experience under my belt I can say that it should have taught me some different lessons:

Persevere in the face of challenges – there are always challenges.

Pivot if necessary.

Success is a mental game as much as it is about execution.

What seems like bad timing might just be the natural course of competition and innovation required to stay ahead.

Hopefully you found this story entertaining. If so, let me know. Thanks for reading.

Powered by beehiiv

March 18, 2024