More specifically, I realized that the technical foundations of OSR+ are now almost four years old (if you don't count the beta site that was in place when we first started)! In Web terms, that's like a decade. And with the pace of genAI development, that's like 100 years.

Why does this matter to the development of a tabletop roleplaying game? It's a pen-and-paper product, right? Who cares if robots take over the world, when the game is in our imagination?

Well sure, that's the correct line of thinking if the only thing we planned to put out were physical books (and don't get me wrong: physical books are the pinnacle of the artform in my mind). But because I come from a technical background as much as I come from a literary background, I believe digital tools are as integral to the development of OSR+ as the physical books. We do the vast majority of our playtesting remotely. We use the Character Creator and its GM tools for almost all of our games, and the website is a critical reference tool behind that. Not to mention the behind the scenes tools we've developed, such as our "Taskmaster" recording stack (a combination of custom web-based tools, OBS, and VDO Ninja).

And there are so many tools we have in the pipeline to create that will ultimately rely on this website being the central hub for their operation.

Which means I can't let the technical foundations get out of date. We need better architecture, that's easy to extend and highly future-compatible, so that we can roll out the enormous backlog of digital tools quickly, at high quality, and safely.

(Follow us on Instagram, Tiktok, or Facebook for more fun animated clips. Or subscribe to our YouTube to follow That RPG Podcast.)

The Art of Agentic Programming

In my day job as a web designer, the paradigm shift that has occurred in the past couple years with genAI has made it possible to produce super complex technical infrastructure at scale, in ways that were literally impossible before. And I'm not talking about "vibe coding" (the dev version of AI art "slop"), where you're just kinda telling the AI to make you things that you have absolutely no understanding as to how they work, and the output of which you never review.

Agentic programming, on the contrary, is like having a bevy of junior developers at your disposal who can help you perform mundane or tedious tasks, so you can focus on high level architecture. And if you're fastidious, you can review all the code your little robot interns produce line by line to make sure they haven't snuck in nonsense.

Case in point: since I shifted gears in March to focus on rebuilding our foundations, I've refactored in two months what took me two years to write in the first place. The new infrastructure is not only super performant, but will be performant at scale, and opens the doors to all sorts of digital tools that will make GMs' lives easier and players' experience more exciting, whether you're playing remotely or in person at the table.

I'm too busy to get in the weeds about specifics just yet, but this post serves as a quick note to say: HEY! I'm still alive! I'm building something bigger and better that will blow your socks off. Just give me a couple months to get there.

What this means in practical terms:

• I've put the Bestiary on hold while I focus on rebuilding the site's infrastructure. The plan is still to incorporate it in 2026, but in the site's new architecture. (If you sign up to our Discord, however, you can get a preview of some monsters from the Bestiary to date.)
• Print is pushed back to 2027 at the earliest.
• The digital tools in our roadmap will come A LOT earlier than anticipated, because once the new infrastructure is in place, the agentic model enables me to rapidly extend the platform to create tools in a tiny fraction of the time it would have taken without it.

Anyhow, that's all I got for now.

In the meantime, come visit us in the Discord (you can even play some games!) or catch us at PAX Unplugged 2026.

...Now I've got to get back to work.