writing through the noise

I'm Jason, and I'm overwhelmed.

I am an engineer who builds software for a living. I consider myself relatively "in the know" about the latest and greatest in tech, especially around AI and its impact on software engineering. And I'm experiencing a vast & complex range of emotions in response to what's been happening lately - emotions that include, but are not limited to:

• Excitement
• Anxiety
• Empowerment
• Dread
• Joy
• Jealousy
• Wonder
• Fear

From my conversations with other engineers at work and beyond, I know that others are experiencing something quite similar. It's pretty intense!

a certain uncertainty

Looking at software engineering a few months down the line, the only thing that I can be certain about is that I can't be certain about anything. Beyond the promise of what exists today (and it's a lot of promise), this uncertainty is underscored by the very real chance that a fresh and exciting launch announcement comes through which changes everything, and creates yet another "before and after" moment for our industry.

Somehow, these moments tend to feel both surprising, and yet, inevitable. And their frequency just keeps increasing.

The blistering pace of innovation bursts open our imaginations with new possibilities, while paradoxically clouding our ability to confidently anticipate our future, making it really hard to make sense of everything that's going on.

At first, playing around with AI in software engineering was fun because it was novel and exciting (which it still is!). But lately, something has shifted significantly for me.

the addictive loop

Once you start diving into this space - if you're anything like me - the addictive feedback loop sucks you in deep. The dopamine hits that come along the way as you prompt entire systems into existence, building what you never thought you'd have time to build, or that you thought were simply impossible for you to build on your own, one more prompt after another until it's been hours and you've forgotten to eat, brush your teeth, or use the bathroom.

Like any addiction, its meaning and purpose eventually become lost in the flow of it all. It's ridiculously fun until it's not, and, personally, I've felt that when I reach that stage, I don't necessarily stop - I just become fueled more by anxiety about the future of my career than I am by any coherent strategy for building something useful for myself or for others.

building in public

It's all too easy to fall into a pattern of building for the sake of building, which, while starting as something I genuinely enjoy, leads to an addictive loop whose joy has diminishing returns. AI & agentic coding tools have dramatically amplified this trend for me.

I often find myself in conversations with my team about, arguably, the most important parts of the software development lifecycle (henceforth SDLC), which I call "the hard parts". These represent everything around the production of code - the research, sales & marketing (even if it's internal!), deployment, productionization, instrumentation, project management, and the upkeep of a well-oiled delivery pipeline. It's stuff that's not typically taught in the classroom, but it's what separates great engineering teams from the good.

My personal projects have essentially been "build-only" - they're missing the key pieces of delivery that actually make software useful. As a result, I've balked at discussing & working on my projects in the full view of my network for fear that they need to be "perfect" for anyone to set their eyes upon them. They have served me extremely well as learning exercises and the one-off tool here and there, but I haven't pushed any of them past that point.

And, critically, as AI & agentic coding tools continue to mature and gain prominence, recent trends in software advancement have convinced me that the "building" part of software delivery will become less and less important when aiming to create real economic value with technology. Notably, the sheer volume of technology products that have popped up in the last couple of years prove that it's easier than ever to build software at a record-low cost. The hard parts of building software businesses - which require taste, judgment, strategy, and human connection to get them right - are where the actual value is concentrating. And, not so coincidentally, I want to get better at them!

so, why write?

Which brings me to the point: I'm going to use this space on my personal website to write about my experiences, ideas, plans, emotions, and other aspects of my engineering profession as the world keeps changing. I'll use it to deep dive into projects, ruminate on ideas, and, for now, try to make sense of all the chaos around us.

It will also serve as a forum for me to work through some of the hard parts of my own SDLC, which I'm hoping will hold me accountable to the high standard of output that I am in continuous pursuit of as my career progresses.

But Jason, why not just start a Substack?

An astute question, dear reader. While I do like writing, I don't like the idea of being beholden to the creation of "content" that needs to be tuned to the preferences of an audience. That said, I might write Substack-y content here from time to time! But the act of putting my thoughts here - in my own little corner of the internet - is designed to give me senses of ownership, accountability, and agency that are purely selfish in their nature. Who knows, in a few months or years time I might change my mind - if so, consider this a disclaimer :).

I also happen to think that I have some ideas that some folks might find interesting, and I want to put them out there to both refine them and to start meaningful discourse (if I'm lucky). The only marketing I'll do (for now) is a LinkedIn post here and there directing my network to something I've written.

If you've made it this far, I hope you find this post and my future writing useful, inspiring, thought-provoking, or downright disagreeable. Come find me on LinkedIn or shoot me an email - I'd love to hear what you think.