Gette Garage: Why this? Why Now?
Building things - Beginnings
Creative Playground
I'm at a crossroads in my career and creative life. After spending years accumulating experience across different domains, I've realized something important about myself: I don't like to be boxed in. While deep specialties have their place, I'm the person who thrives on tinkering, exploring, and connecting dots across disciplines.
Right now, I'm making a game. It's a solo project that lets me combine my love for drawing, music, and programming. It’s an ambitious challenge, but the perfect creative playground to explore how AI tools can help solo developers like me tackle big challenges.
I also like woodworking, fixing old motorcycles, and hiking but I haven't yet found how all of them fit into my ideal creative workspace. Maybe there's an ikigai connection here (finding purpose at the intersection of what I love, what I’m good at, and what the world needs). Then again I can't help but think of Steve Martin's The Jerk and his search for a "special purpose", so for now I'll roll with what I have.
Beyond being a creator, I'm also a data scientist and AI Engineer. Recently AI tools have exploded onto the scene, and I see immense potential in how they could transform game development - particularly for solo devs and small teams where every efficiency gain matters. AI can potentially assist with animation, dialogue, and brainstorming, but it also presents creative challenges. Games are an art form, and while AI can generate assets or even write code, the human touch and details are what give a game its soul.
So why write a blog? I've never been much for social media's quick takes and engagement algorithms. I prefer the control and depth of long-form content where ideas can develop. This Substack gives me a place to document my journey, share my findings, and hopefully connect with others.
Me So Far
My background is maybe a little unconventional for a game developer. I spent years in academia, eventually earning a PhD in physics with a Master's in astrophysics along the way. After academia I moved into data science where I've spent the past six years focused in product development. This taught me how to translate ambiguous ideas into tangible tools that solve real problems. I’ve worked in both startups (Edison Software) and government (U.S. Office of Personnel Management), giving me experience with different work environments, constraints, and development cycles.
No matter where I go, I keep coming back to games. While at Edison Software, I built a gaming analysis product that tracked and visualized unique data trends in the industry. Things like wishlists, pre-orders, release spikes, and the long-tail effect after launch all tell a story about a game's trajectory.
At OPM, I led a team of 15 in the AI Incubator where we explored and prototyped new AI-driven products. We looked at a variety of solutions ranging from document parsing to skills matching. One of the biggest challenges we faced was uncertainty: "Is this AI product actually going to work?" So I borrowed a concept from game development: rapid prototyping. Like a game jam, we created design documents and small, interactive "playable" prototypes to test ideas quickly. This approach helped us iterate faster, cut through ambiguity, and get to working solutions in weeks instead of months.
I was in Human Resource Solutions, specifically in the Talent Acquisition Analytics Group. After three years, I still have to double-check whether it’s 'Talent Analytics' or 'Talent Acquisition.'
Now, as I build my own game as a solo developer, I'm confronting the practical challenge of needing to do everything myself. This constraint forces creativity. For instance, while I'm comfortable with some traditional media like charcoal and oils, pixel art and sprite animation for games require similar, but different skills. This is where I see AI tools potentially helping with that workflow. Tools like Stable Diffusion may help me iterate on character designs or generate animation frames that I can then refine. LLMs can help brainstorm new features, characters, or write dialogue. It's not about replacing the creative process, but rather augmenting my workflow, filling in gaps, and letting me focus on the aspects of development that matter most.
My Goals
First and foremost, I want to create something valuable for the game development and AI communities. There's a lot of noise out there, and my aim is to contribute practical applications, real experiences, and honest assessments of what is working and what doesn't.
For myself, this Substack serves as a repository of my journey. Game development is a long and iterative process, and documenting it helps me track progress, reflect on lessons learned, and stay motivated. There’s something powerful about looking back and seeing how far you’ve come.
I'm also committed to continuous learning. By sharing my process publicly, I'm creating accountability for myself to keep pushing forward, trying new approaches, and applying what I know to practical challenges. Writing about what I’m building forces me to think more deeply about my decisions, and hopefully, those insights will be useful to others as well.
You can expect a mix of content here at Gette Garage:
Game development logs: Regular updates on my game's progress, challenges I'm facing, and solutions I'm discovering
AI experiments in gaming: Hands-on tests and explorations of how AI can assist (or complicate) game development
Project showcases: Deep dives into side projects like VineScribe (my AI wine companion) and, of course, my game
Technical insights: Lessons learned, code approaches, and architectural decisions that might help others
I want this to be a workshop where ideas are built in public. Sometimes it will be messy, but it will always honest, and hopefully useful to others walking similar paths.
Finally, I hope to build a community of like-minded creators and people interested in the intersection of games and AI, who want to share ideas, provide feedback, and perhaps even collaborate on future projects.
What You Can Expect
I'm committing to a bi-weekly publishing schedule. Sort of like a sprint review for those familiar with agile development. This cadence gives me enough time to make meaningful progress between posts while keeping a steady rhythm of updates.
The content will blend casual updates with deeper technical dives. Some posts might be quick peeks at what I'm working on, while others will thoroughly explore experiments or concepts I'm testing.
Regarding technical depth, I'll start more accessible, but I'm willing to go as deep as readers would like. I’m comfortable with complexity but value clarity and real-world application over theory. The goal is to keep things insightful and engaging.
Who are you?
I'd love to hear from you! Who are you? What brings you here? Are you a developer, an AI enthusiast, a gamer, or simply curious about this intersection of technologies?
What aspects of game development or AI applications are you most interested in learning about? Do you have specific challenges in your own projects that you'd like to see addressed here?
Feel free to leave comments on this post or reach out directly via email. I also may set up a Discord server for more real-time discussions and community building as this grows.
This is just the beginning of what I hope will be a valuable resource and community for those of us exploring the frontier where games and AI meet. Thanks for joining me on this journey and let's build something!