Our Start Up Curve


If I thought 2021 was bad, 2022 was probably worse. However, after surviving 2021, I was better prepared to deal with the 2022 emotional rollercoaster. 

At the beginning of 2022, we had these objectives:

    • Raise another round (we were running out of cash)
    • Finish the lamp design
    • Build, sell, and ship 50 Beta units
    • Start sourcing the plastic injection molds

Well, the year started out rough. We spoke with more than 200 potential investors and were not getting traction. Either the investors had issues with the cannabis component of our solution and passed OR they were into the cannabis component but gave us terms that were somewhat predatory. In other words..... really bad terms.

3D Printing

We also had the stress of supporting our alpha units from 2021 and printing 50 Beta units. To put things in perspective, making each machine took us roughly seven days of non-stop printing. This was a massive effort. We invited all sorts of friends to come over and help with printing, building, testing, packing, etc. It was a ton of manual labor and a lot of long days, including weekends. 

By the middle of the summer, we had very little money and no real traction with any VC. Things were starting to look dire and I had to make a difficult decision. The company was able to keep going, but we would have to cut our team entirely. I started suggesting that our team members look for jobs elsewhere because we were going to be completely out of cash by October. 

By November, I was the last man standing and had $7K in the bank.

Well, that’s not entirely true. Boundless Robotics was lucky enough to be surrounded by an incredible group of advisors, friends, family, alpha testers, you name it. Despite not being able to pay people, everyone pitched in, including the former engineers who were now looking for another job or already at another job. Everyone wanted to see this through and see it succeed.

First Annabotos

We began the year with a few alpha testers in the field and were starting to see how well our product actually worked. The lamps we had designed with Target Salad Bowls (we named them the pharaoh lights) were working really well and we were making strides in the design and development of what would become the final lamp; the tripod!

Then, an alpha tester went on Reddit and posted a picture of his Annaboto with a plant in it. This caused a lot of excitement and within a day we had found ~100 people in Massachusetts who wanted to become Beta Testers AND PAY FOR THE ANNABOTO! This was very exciting. People really wanted to purchase our product!

New lamp

When it was time to deliver the systems to our beta testers, we had very little money so we could not afford to ship them. Instead, we invited them all to our office to pick up their machines. We bought a bunch of beer and chips, gave them a tour of the printers, and were able to demo how to build the product and how it worked. This ended up being an AWESOME event because we were able to  personally get to know our diverse and loyal customer community who have been instrumental in our success.

TechCrunch Disrupt

We also learned that we had been selected as finalists for Tech Crunch Disrupt. Finally, we were going to have our Silicon Valley Show moment! AND, we finally found a VC (Boost) who was actually interested in investing. Since they are California based, I was going to do Tech Crunch and my final pitch with Boost during the same week. I almost canceled because I couldn’t afford to fly to SFO, but a friend went ahead and purchased tickets for my wife and me. Yes, my fabulous wife put on her Boundless Robotics hat and accompanied me to staff the booth at the show! And that friend who bought the tickets also flew out with us to provide support and encouragement during Tech Crunch Disrupt.

Annaboto in your living space

In addition to all of this, we had launched an equity crowdfunding campaign and were actually beating our goal of raising $250,000. People were very excited about Annaboto.

By December, we had 50 active systems in the field with mostly happy customers, we had closed an investment by Boost VC (this turned out to be SOOO much better than expected - Boost is incredible!) and we raised ~$400k through the crowdfunding. With this money, we were finally ready to start scaling our production and help people effortlessly and stylishly grow cannabis at home!

The year started out rough and we had to make serious cost cuts in order to keep the company alive. In the end, we were able to raise close to $1.2M through Equity Crowdfunding and Pitching to 200+ investors. A serious question remained: how was I, the only full time person, going to be able to scale up manufacturing, update the software and mobile app, and start our marketing efforts? 

Then ChatGPT came out.

Lessons Learned:

  1. Your reputation and network matters more than you think: I have always worked hard and people know that. I also always help when I can. And people know that. When I needed help, everyone showed up (I’m getting teary eyed).
  2. There’s more than one way to fund a company: F&F rounds, VC, Angels, Equity Crowdfunding. Get creative. Know your investors. Pitch accordingly. 
  3. Never, ever give up: I believe in the product. I could have given up after my 100th “no”. I could have given up when we had no more money. Things came together in the end. They always do. 

What to Do Differently:

  • Understand the investor before you reach out: I didn’t really understand that there are sweet spots for investors. Duh. Makes total sense. I was trying to push a hardware consumer product for cannabis to people who invest in software and B2B plays. That was a waste of time. Also, it turns out that you can do a lot of research on investors and know whether they will be a good fit before you even reach out. I realized too late what type of company we are and what type of investor we need.
  • Should not have considered China as a manufacturing option: Because this is a consumer product, we thought China would be a good place to make it. This was a bad assumption because of the amount of work needed to get it done and the lack of people at Boundless to do it. We went local because that way we could mitigate more risks in order to get us going. We can reconsider this in the future. 
Back to blog