Welcome to the Tip Sheet for APIs, AI, and Data Science. This week I am going to share some advice for keeping focused on learning when the hype cycle gets overwhelming, as it was last week. We'll also meet a new expert coding in the open, and see a blog post on using an API Gateway with FastAPI.

Keeping Your Head in the Tech Storm

When you see the technology and business world go wild about the Next Big Thing, how do you react? Excitement? Anxiety? FOMO? Fatigue?

To begin with, I hope that you're learning, building things, and growing your skills. If you're doing that, by default you're better prepared to handle the hype.

But even if you are doing those things, public enthusiasm over new technology and products can be a double-edged sword. When a brand-new category of tools or technology is announced that has a lot of value or potential, it can be worthwhile to dive in and start learning it. There will often be a lot of resources that become available, and you'll be learning along with the industry, which is a great chance to share your experiments and projects through blog posts, code repos, social media posts, and the like.

But sometimes the pace of change and level of hype is so great that it becomes a bit overwhelming. The number of options of things to learn becomes so large that it's hard to know which ones to focus on. And with each new big release, you wonder: Should I stop learning or building what you have in progress and focus on the next?

This week (February 2025), I've observed that dynamic occurring with the announcement of the DeepSeek AI model, but with the current pace of things, there always seems to be another Major Event waiting in the wings. Here's a bit of advice for how you can make consistent progress in your development without letting this distract you.

Mindset Tips

My first few tips are related to mindset, planning, and perspective:

• Become a finisher - Completing the project you're working on is more valuable than starting a new one. Focus on finishing.
• Understand your goal - Take a few minutes to consider what you are trying to accomplish in your technical learning. If your development goals are directly tied to your current job, it will be a bit simpler, because you can just focus on tools that you're currently using or will be soon. If you are working on personal development, consider what you plan to do with the technology. More than likely the latest advance that gets all the publicity won't change this goal much, so you can filter it out. (On the other hand, if your goal is to stay up to date on the latest water-cooler chatter, you may enjoy reading the Latest Big Thing.)
• Plan in the medium-term - Set your learning goals in the 1-2 year range. Consider where you want to be at that point, what skills you want to have, and what deliverables you want to be able to show.
• Control the FOMO - With your goals in mind, and a medium-term perspective, you should be able to control the fear of missing out on the latest big hype. Meeting your goals is enough -- you don't have to learn everything at once!
• Avoid Islands of Isolation - An item that has stuck with me from my Java programming days is a concept called Islands of Isolation. When the Java garbage collector found objects that only referenced each other but weren't referenced by anything else, it could clean up their memory (I'm probably over-simplifying the concept). That image has always stuck with me as an analogy in life: some news stories or technologies generate a cycle of hype and response that take on a life of their own, but in time are superseded by other events. In the medium term, they may not affect me at all, they don't need to pull me away from my work.

Hands-on Tips

In addition to the tips above, here are some more practical ways to address the hype cycle:

• Build a base project - For your learning goals, build a base project that you can enhance and evolve with related technologies. For my book, I started with a basic FastAPI API and then built additional items on top of it (SDKs, Streamlit Apps, etc.) If you start building a base project, you can use it to test out any new interesting technology that comes on the scene. Then create a blog on it and share your learning. If it's valuable, maybe you keep working with it. If it doesn't provide any new value, you move on.
• Focus on a specific industry or domain - As you do your learning or side projects, narrow your focus by domain. Most of my side projects focus on fantasy (American) football. So I always ask the question: how does this new thing address fantasy football? A great example of this in the business world is Brent Lemieux, the owner of a data analytics company that focuses on lawn and landscape businesses. Here's a post about using data to grow leaf cleanup services. When you have a focus like this, you have a great funnel for the Next Big Thing: how does it improve what I'm doing in my focus domain? How would this impact my focus domain?

Meet the Experts: Kade Halabuza, Software Developer

One of the most enjoyable parts of writing my book was learning from all the cool work that others were doing around APIs, AI, and Data Science. While I was working on my chapter on ChatGPT, I came across several articles about Custom GPTs written by Kade Halabuza, a freelance software developer. Kade tried out Custom GPTs with the goal of creating a Hockey Stats Expert (did I mention Kade's from Saskatchewan, CA?)

He first wrote an introductory blog post about creating Custom GPT with step-by-step instructions. Then he followed it up with a second blog post showing a custom API wrapper he created to address some of the limitations of the original GPT. He also created a public github repo with the code he used.

If you're interested in Kade's latest work, he is sharing SEO advice from his Starter SEO Audit service. Here's one of his recent posts on boosting click-through rates. If you want to see a great example of building and sharing, check out Kade's work.

A Starter API Gateway for Your Side Projects

There's a new blog post this week written by Marcelo Tryesinksi, one of the main contributors to the FastAPI project. It shows how to use the Zuplo service as an API gateway for an API you host on Render. I haven't used the service, but it looks like an interesting way to check out some features like rate limiting and authentication for a side-project API. Here's the article: FastAPI Tutorial: Build, Deploy, and Secure an API for Free.​

That's a wrap for this week!

Keep coding,

Ryan Day

​https://handsonapibook.com/​