I started as a Digital Artist, built production pipelines for virtual production stages, and now I architect AI-powered developer infrastructure in Rust.

10+ years across the full arc, from creative tools and VFX pipelines to fullstack development, systems engineering, and AI agent orchestration. I build the systems that let small teams ship like large ones.

Built key production tools for Eyeline Studios' virtual production stages, including LED volumes, volumetric capture, and the Light Dome.

• AI agent infrastructure: autonomous dev agents that go from issue to merged PR
• Code intelligence: giving AI agents structural understanding of codebases before they write code
• Developer tooling: workflow orchestration, self-healing systems, reducing friction at scale

I built and operate an autonomous agent network that functions as a one-person software team. I create GitHub issues, and the network handles PRD writing, implementation, CI fixes, and review responses end to end without intervention.

Built on a heavily customized Spacebot fork (846+ commits) with a dynamic project registry, GitHub App with org-wide webhooks via Tailscale Funnel (zero open ports), and isolated git worktrees for parallel execution. The CI/CD pipeline alone has processed 7,300+ automated task commits since January 2025.

• Auto-generates structured PRDs before implementation begins
• Complexity-aware orchestration where simple tasks get one worker while complex tasks get a planner and implementer
• Full PR lifecycle management: CI failure → auto-fix, review feedback → auto-revision
• Cross-channel coordination: GitHub webhooks, Telegram notifications, worker management

Stack: Spacebot (Rust) · Claude Opus/Haiku · Gemini Flash · Tailscale Funnel · GitHub App · SQLite · systemd

A code intelligence engine in Rust that parses source code into a queryable knowledge graph, so AI agents understand architecture before they write a single line. Published on crates.io.

Instead of letting agents grep through files and guess at structure, Myceliums gives them a structured graph of symbols, call chains, functional modules, and execution flows, all queryable through MCP, a CLI, or a custom Cypher dialect. One search call instead of fifty.

• 5-crate Rust workspace (~8,600 lines) covering core, storage, MCP server, Cypher engine, and benchmarks
• Parses TypeScript, JavaScript, and Python via tree-sitter with two-pass symbol resolution
• Hybrid search: BM25 keyword + vector semantic + Reciprocal Rank Fusion
• Impact analysis: parses git diffs, then BFS through the call graph to find blast radius
• MCP server with 11 tools and 8 resources for direct AI agent integration

Stack: Rust · tree-sitter · LanceDB · fastembed · graphrs · rmcp · Cypher dialect