What Was It Actually Like to Build a Production Application?
The Mission: Empowering an Entrepreneur
Tenkiller Hideouts is an application for small businesses to rent and manage a portfolio of properties without having to use apps like AirBnB or VRBO. The overall need supported in this endeavor was to help an entrepreneur scale, it's that simple. The client wanted something simple to use but with all the back-end functions for cyber security.
Starting with a Blank Slate
Let's level-set on expectations before we get too far. To this, I'd say there were no expectations… by this, I didn't know what I didn't know, which was beneficial for two main reasons. First, I had no bias of preference or "the way it's always been done" mentality that hinders a lot of projects. Second, we really wanted to see if what we're selling actually works as advertised.
While describing my skills as having zero development experience is true, I have had many opportunities in my career integrating new systems, workflows, and overall data management. It should go without saying, I was extremely aware of what the objective was, and going in with an open mind was crucial to my success.
Overcoming the Builder's Anxiety
I was very skeptical that a person with no development experience could actually build an app. What happens if I get it wrong? How long is this going to take to fix? I've heard countless horror stories where a person had to rework the requirements and it added weeks or months to the timeline… I didn't want that to be me.
These were some of my initial concerns as I faced the core phases of the build:
- Requirements generation and refinement
- Establishing architecture and actual coding
- Specifications traceability
- Functional testing
- The final mile
As it turned out, those fears were no big deal.
The Beauty of Breaking Things (and Bouncing Back)
The Alchemist product really does account for everything you said you need… but, if you miss something—which I did—it automatically updated the epics and use cases based on the changes made through "Add Insight."
I was afraid to 'break' something and in the course of breaking things (oh yeah, it happened), I learned very quickly that failure is what creates the good stuff. What did not surprise me was that human input is still VERY much required to go from requirements to a usable application. There are too many fine details that only a human can answer, but as I plodded through the plan, my confidence increased as well as the viability of the application I was building.
The 38-Hour Turnaround
I would say the proof is in the pudding… in less than a work week (38 hours), I was able to download all the programs I needed, understand what I needed to do, and then did it. In the end, I had a fully functioning tool where I was able to trace back every function to the initial requirements. Of all the times working integration, this has never happened before.
What I Tell Anyone Who Will Listen
Here's what I tell my colleagues and really, anyone who will listen to my experience:
- Requirements quality determines project quality. If there is any confusion about what is needed, you will waste time.
- AI accelerates implementation but doesn't replace ownership. While I can say that without AI, my task would not have been possible with my skill set… however, AI couldn't have recreated what I did without me.
- If you think you know the best way to do something, chances are you don't. Don't be afraid of taking a new path to accomplish what you know. Finding new ways to do something tends to support other actions down the road, and the more we learn, the more capable we are of solving other, potentially more complex problems.
The Skeleton Key to App Building: Would I Do It Again?
This is really one of the easiest questions I've ever pondered. 100% yes.
Having the ability to take an idea and executing with success is one of the greatest feelings out there. Understanding the process and having gone through this experience has truly opened the door of possibility in a way that is indescribable. Past experience has provided numerous occasions where systems were replaced with something worse. I feel like there are tons of great ideas out there, but no one has really developed the skeleton key until now. Literally anyone who can navigate a computer can do what I did, and I'm still shocked when I hear colleagues say they haven't used it yet.
