After working for a few years in the space program as an aerospace engineer, Joshua saw a need and was inspired to create CADWOLF, the “Google Docs of Collaborative Engineering”.
Davis Baer: What is your background and what are you working on, and what motivated you to start your company?
Joshua Foxworth: My name is Joshua Foxworth and I finished college in 2005 with a BS and MS in Aerospace Engineering from the University of Texas at Austin. After that, I moved to the Houston area to work in the space program where I worked on the shuttle program for a few years and then designed components for the Orion capsule.
It was in this capacity that I realized that something was askew with how engineers approach large projects. While modern CAD has replaced hand drawings and analysis tools have greatly advanced our capabilities, the general process of engineering remains the same as it did at the onset of the space program. Program heads develop overarching criteria that eventually trickle down to hard requirements for engineers designing components. Those engineers then perform their calculations in Excel, Matlab, and several other tools, write documents to explain their results, draw CAD, and then track items like weight, cost, and safety factors in spreadsheets that are always out of date. Each of these facets remains divorced from the others, each has to be largely redone when something changes, and none of this is accessible by management.
Eventually, I decided to take a crack at changing this process and CADWOLF is the result of that work. The platform lets users create individual web pages that have the look and feel of a Word or Google document, but have the ability to solve equations, make charts, create free body diagrams, and do a number of other things. Our part tree system then lets users build large structures on a component by component basis and track items like weight in real time without the need for separate spreadsheets. All of this links to the Onshape CAD system meaning that the CAD model will always reflect the current design math and items like weight and volume can be read from the designs in real time as well.
One of our standard documents with sliders, text, equations, and charts
The end result of this is a system where engineers design their components one time and then those designs update automatically to changing requirements. The documentation, design math, and current status of each part is available in real time. This will drastically reduce the amount of time and resources needed to build items like space capsules, bridges, drilling platforms, etc.
A part tree where users build a system piece by piece and get real time feedback
What went into building the initial product?
While I had experience writing some software, the languages generally used by engineers are not really similar to those used in web development. I knew some PHP from a previous project, so I started there and learned the Laravel framework. I then learned javascript and eventually decided to use the angular framework. Before too long, I was deployed on heroku and using a number of standard web tools like gulp, npm, and git. I also learned Python for a later facet of the platform.
While the goal of the project is to eventually encompass the entire engineering sphere, including things like finite element analysis and computational fluid dynamics, I decided to start with the goal of simple documents that solved and displayed mathematics while sharing data between documents. The hope was that this would provide enough financial backing to develop the remaining functionality.
When this was complete, I put out an MVP. In a field that is slow to adopt new technologies, the MVP just didn’t have enough benefit to convince people to start using it.
This began to change when the Onshape CAD system was introduced. They provided APIs that finally allowed CADWOLF to link our math directly to the design parameters of their parametric CAD system.
Once this was done, we wrote a python script containing the NumPy library that users can run on their personal computers to solve the equations in the browser. This greatly increased both performance and the complexity of math we could offer.
During this time, I also had two children so the code writing had to take place at night and on the weekends. You’d be surprised by what was possible when sleep is no longer a factor, but the entire process still took about 4 years from start to where it is now.
How have you attracted users and grown your company?
There is a well established method of introducing software to the engineering community. While the system is directly marketed to engineers, it is heavily promoted to university students free of charge. Eventually, those students will trickle out to the workforce and want to continue using that platform in their professional careers.
To get in front of students, we offer help with homework on our subreddit and maintain a bank of homework problems in several subjects. This introduces those students to the basic functionalities of the system with the hope that they will be comfortable enough with the basics to use the higher functions when the time comes.
We also offer free licenses to organizations like Engineers Without Borders or any other charitable group.
In the coming months, we plan to start organizing projects on the platform centered around everything from building a flugtag, to designing a drinking well and pump, and eventually spacecraft ideals. There is a real need for this sort of community in the engineering field — especially among young engineers who lack mentors with experience.
What’s your business model, and how have you grown your revenue?
The business model for this type of system is very straight forward — a monthly or yearly recurring license fee. However, there are two ways where we differ from the model of other engineering platforms.
The first is that they make everyone pay for a license and/or severely restrict student versions. This prevents potential users from investigating and learning the system. It is a barrier to entry between you and your users.
The second mistake is that they charge significant fees for training. Again, the goal should be to provide opportunities to learn the system and not barriers to accessing it.
Of course, this limits your revenue so you have to go into the field with the understanding that engineering moves very slowly. Our model has been to charge users if they are using the platform in a professional manner to make money. The long term goal is to change the inertia of engineering community so that our platform eventually becomes the default.
What are your goals for the future?
With roughly 10 to 15 million engineers in the world and millions more who could use the system for its mathematics and organizational abilities, the market has significant revenue potential. However, the goal is to alter the way engineers approach their work.
When a web developer begins a project, they can build upon the functionality that already exists in frameworks and then bring in additional packages through something like npm. In engineering, every project starts from scratch and every engineer has to design and redesign every part. With CADWOLF, engineers will be able to pull in parts from a bank of predesigned components and have those designs adjust to the given loading conditions. Engineers will be able to focus on designing the items that are unique to their system and not reinvent the wheel over and over for each project and redesign.
Aside from this, I really want to see CADWOLF used to develop a space vehicle.
Have you found anything particularly helpful or advantageous?
I think that for people that have been doing startups for a long time, there is a tendency to forget how great the community is around startups. Everyone is rooting for everyone else to succeed. People are usually very giving with their time and advice, and many will work to help you wherever possible. I haven’t really encountered this in any other work environment. It’s very uplifting.
What’s your advice for entrepreneurs who are just starting out?
If you are a technical person, don’t just tackle the problems you know. There are a lot of very talented coders in their early 20s who are tackling issues like parking, or bike sharing, or hiring. The reason for this is that you can’t solve a problem that you are unaware of, and if you are 23 years old the bulk of your pain points have revolved around college, parking, social media, and hiring. Instead, meet people who have worked in outside fields and decided that those fields are in need of innovation.
Where can we learn more?
The system is live now. From there, you can get to the documentation folder and read through how the system works. We also have some tutorial videos and are always increasing our bank of homework problems.
If you have any other questions about the future of the platform, the future of engineering, or the future of the space program, just ask me below.