BLAKE
WERLINGER
About
Self-taught,
AI-obsessed.
Developer Journey
I got my start with Scratch around age 11. Nothing fancy, just dragging blocks around trying to make a clicker game that didn't suck. I spent way too many hours on it, but that's where I learned the basics: loops, conditionals, the idea that you could tell a computer what to do and it would actually listen.
By 14, I had moved to Unity and C#. My first real project was a top-down zombie shooter that I was convinced would be the next big indie hit. It wasn't. But I learned more from that half-finished game than I did from any tutorial.
That's kind of been my pattern ever since: I learn fastest when I'm building something I actually want to exist. I don't learn well from tutorials, I'm self taught because I have spent many sleepless nights fixing something that I had the vision for, but had to develop the skills required to build it.
Since 2018, I've been active in various developer communities. I've made mods for games like RimWorld and Valheim, built websites for small businesses, prototyped game demos that never saw the light of day, and contributed to open source projects when something caught my interest. I like variety. I like solving problems I haven't seen before.
If I'm being honest, I love the early stages of a project almost as much as finishing it. There's something about sketching out an architecture, debating trade-offs, and figuring out how all the pieces fit together that just clicks for me.
AI Obsession
I've been fascinated by AI since before GPT-3.5 dropped in late 2022. The idea of machines that could reason, even in a limited way, always felt like science fiction becoming real. When LLMs started getting good, I was hooked.
Back in my senior year of high school, I taught myself Python specifically to build a "Jarvis" style voice assistant. This was pre-LLM, so I was cobbling together speech recognition libraries with a basic PyTorch NLP model that matched my voice commands to a predefined list of actions. It barely worked, but I learned a ton about how these systems are actually built under the hood.
In April 2023, I got the chance to work with Toran Bruce Richards and other early contributors on AutoGPT. This was during the project's explosive early growth, when we were all figuring out what agentic AI could even look like. I contributed code, participated in architecture discussions, and helped shape some of the early thinking around how autonomous agents should handle tool use and memory.
My demo video was featured in AutoGPT's official README for several months during development. Despite the rough audio quality from my laptop mic.
AutoGPT ended up becoming a landmark project—it's still the 24th most starred repository on GitHub. My contributions weren't massive in terms of lines of code, but being in those early conversations about the future of AI agents shaped how I think about this space. I'm still genuinely passionate about Agentic systems like AutoGPT, Claude Code, Cursor, and other tools that are pushing the boundaries of what AI can do.
Specialty
AI Agent Systems
Planning architectures, tool orchestration, context management
Core Stack
C# / .NET
Backends, game engines, high-performance systems
Also Fluent
Python, TypeScript
AI tooling, web apps, rapid prototyping
Approach
Iteration is Innovation
Same concept, different contexts. That's how you find novel solutions.
Contact
Let's talk.
I answer emails and DMs fast. If you want help on a system, want feedback on an idea, or just want to swap notes on AI tools, reach out.