The primary subject of my blogs is tokenization and related topics, but every now and then, I like to write about my experience as a Project Manager. I often think about creating a separate account just for that kind of post, and I may do that one day, but for now, I'll just use this one.
One of the things I take pride in as a Project Manager is diligent documentation. A common complaint I see in this industry, especially in tech, is that projects are often poorly documented if there is any documentation at all, but not my projects. If anything, I "over-document" and I'm alright with that.
I have this project that I've been leading for over three years now. It's a support project with the goal of optimizing the performance, SEO score, and accessibility of dozens of clients' websites across the Americas. Due to the nature of this project, its scope is not always predictable.
We are constantly dealing with custom demands from the site managers to fix very specific problems or implement different features that they want to test. Over the last three years, we have worked with a lot of those requests, declined a few of them and made decisions based on their output.
We have done so many different things now that every now and then, we get a new request from one of the sites, and just by looking at it, I can tell whether we had done something similar before or declined something like it before. The problem is remembering exactly what was or was not done and why.
As I mentioned, I'm very good at documenting this stuff, so I have literally thousands of bytes worth of documentation for this project, which is great, but it also creates the problem of navigating through it.
Even with good structure, naming conventions and smart search functions, it can be tricky to find stuff on the go when you need a quick answer, but recently, I decided to do some experimentation, and the results have been great so far.
I hooked this database to a LLM and I'm using AI to provide answers to prompts such as "Have we done something like this before? What was done? If not, did we specifically recommend against this? Why?"
I still need to tweak the model and the prompts, but it's already helping me immensely. So much so that I decided to expand this to my personal life.
I realized that on many occasions when I'm facing a challenging situation, I know somehow that I've gone through it before or felt like that before, but I cannot recall exactly what the context was, and, more importantly, I can't remember what I did.
In the face of that, why not try a strategy that's been proven to be a successful one in my professional life?
It won't be as easy because I don't have extensive documentation of my personal life, which means I will need to resort to the ancient art of journaling.
I used to keep a journal some time ago, but I stopped keeping track of it, so I'm many years behind. Also, I'm a big fan of handwriting, and that's how I used to do my journaling, meaning whatever little documentation of my own life I have is handwritten and not suitable to be used by AI, so I will need to take some time to convert it to a digital format.
There are some modern tools that may help to streamline this process but I'm seriously considering typing it all manually because that would be an interesting way to "re-live" those memories. I'm not sure where all of this will take me, but I'm definitely looking forward to finding out.
Final thoughts
Keeping thorough documentation of my projects has proven to be a key decision to increase my efficiency with the aid of Artificial Intelligence. Unfortunately, I wasn't as diligent about documenting my own life, so I'll have to start almost from scratch if I want to apply the same strategy to my personal matters.
I know I have a lot of work ahead of me but I'm excited to start this new journey. Not only because I believe this setup will help me to deal with future challenges and better process my feelings, but I'm also looking forward to getting back to journaling. I'm sure it will be a great self-awareness exercise.