9 years in the games industry. Currently spearheading the full automation of 3D art pipelines at Attensi. I use context routing and generative AI to turn repetitive production work into freeing up time for designing and engineering new and better tech.

Unity · C# · Python · Meshy · Claude Code

Spearheading the move from a tool-driven automated pipeline to an AI-native one. The current AI pipeline is built directly on top of the character creator and tooling I designed before the current generation of AI.

• Generative 3D in production. Meshy is integrated into real revenue-generating flows. Hyper3D is next in line, not in yet.
• Image and video synthesis. Hands-on hobby experience with Stable Diffusion and Wan 2.2, combined with concrete plans to bring them into professional pipelines in 2026.
• Building on existing foundations. The character creator and Unity tooling I designed earlier are what made this transition possible in the first place.
• Context-routed art pipelines. Treating art production the same way Klonode treats code: every asset, style guide, and tool has a known place, and AI agents only load what they need for the current task. Cuts noise, keeps outputs on-style, and makes it feasible for a small team to run many parallel AI-assisted jobs without losing coherence. Even with smaller local models.

TypeScript · Svelte · Node.js · Claude Code · ICM

Side project. Klonode auto-generates per-folder CONTEXT.md routing for a whole git repo so AI coding assistants load only the context they actually need. Cuts Claude Code token usage dramatically and lets agents reason over large codebases without drowning in noise.

• Auto-discovery. Language detectors for TS/JS/Python/Go/Rust/PHP/Svelte/Vue and more.
• Dual view. Tree and graph, so humans and agents see the same structure.
• Self-improving. Routing optimizer that watches real agent traffic and refines context boundaries.
• Workstation wrapper. CLI, Web UI, Desktop. Chat directly against the routed project.

LLMs · Unity · Game Dialog Systems

I built the initial product demo for the demo that became RealTalk, integrating LLMs into games to give players genuinely freeform dialog. It became a main part of Attensi's marketing and has since grown into a product selling to big companies. I handed my initial demo to another team that made it into the industry-leading product it is today. I'm proud of both halves of that: proving the idea well enough that the company committed to it, and handing it off cleanly so a specialized team could take it further than I would have alone.

Unity · C# · Custom Editor Tools · Animation Rigging

Earlier in my career, 3D character delivery was labour-intensive. Every client project meant grinding through the same work by hand. I therefore made an industry-leading character creator for Unity that automated the bulk of that pipeline and shipped game characters to some of the largest companies in the world. I was in charge of all the logic, architecture and both 3D technical development and 3D asset production (with help on parts of the tool implementation). It predates current generative AI, though I used narrow AI for small pieces even back then. It's the system that taught me what a world-class character creator actually needs when it has to run well on mobile and web, not just look good in a render.

The reviews for the character creator have been nothing short of amazing. It shipped as a core part of Attensi's Creator toolkit and is genuinely fun to use, which is something customers bring up on their own. Getting a 3D production tool to land as "fun" instead of "tolerable" is something greater engineers and designers have failed on before, and it's one of the things I'm most proud of being able to accomplish.

What has changed now is the cost floor. Generative AI has made entire classes of pipelines viable that simply weren't worth building before. My focus has now shifted and my eye is tuned at how I can use AI to reshape the industry.

I am not a formally educated software engineer, I'm a technical artist who learned to code because the problems in front of me needed solving. 9 years of shipping games taught me how to set up pipelines that can deliver effectively and consistently in large scale productions.

What that means in practice:

• I know my way around C#, but I am not a code reviewer in charge of core and rather specialize on using tools and scripting to build production pipelines.
• I've been using LLMs, diffusion models, and 3D generators since the early days, in pipelines that make real money, not demos.
• I have great intuition for how to deliver successful large scale 3D production pipelines.

Production

Generative AI Stack