Skip to main content
AboutShowcaseLet's Talk

I architect production AI systems.

Agent Orchestration, Distributed Systems, Cloud Infrastructure.

Schedule a CallSee Past Work
Currently buildingMossaic

The Process

1

Architect

Discovery

Read the system end to end. Design the solution. Surface tradeoffs and failure modes in a brief that scopes the build.

Design Brief
Arrow down
2

Build

Implementation

Implement the system end to end. Real architecture, real infrastructure, real error handling. Code that runs in production.

Repo + Tests + Docs
Arrow down
3

Ship

Operation

Deploy, operate, iterate. I keep ownership in production and ship the next thing alongside your team.

Retainer Available

Featured Work

Voice AI

Mossaic

Voice-first research recall built on Moss. Ask your user research out loud, get cited answers in under a second. Dual-surface: public WebRTC voice agent for product teams, Slack bot for researchers, one shared index.

MossLiveKitPythonNext.jsNeon
Explore →
Systems Engineering

Argus

A TUI-based unified control plane for agent sandboxing, written in Rust. 8 modules, 9 dependencies, no async runtime. 3.7 MB binary.

Explore →
Product

ArchByte

Real-time architecture maps for any codebase. Your model, your machine.

Explore →
Showcase

Silicon Cortex

Event-driven AI agent orchestration on Kubernetes. Zero idle cost.

Explore →
Showcase

Meridian

Uber for RoboTaxis. AI agents coordinating at scale.

Explore →

Frequently Asked Questions

What AI systems do you build?Toggle
Agent orchestration layers, event-driven pipelines, multi-agent architectures, and developer tools. Rust for systems work, Python for AI services, TypeScript for full-stack. Every project ships with real error handling, observability, and infrastructure as code.
What is the Customer Zero approach?Toggle
I run every system I build in production before I ship for anyone else. Real workloads, real failure modes, real fixes shipped. By the time a client sees it, the second-mile integration friction is already gone.
What does a Build Sprint look like?Toggle
Production-grade implementation shipped end to end. Real architecture decisions, infrastructure as code, error handling, observability. Code that runs in production, not a slide deck. Delivered as a repo with tests, docs, and a video walkthrough.
What tech stack do you work with?Toggle
Rust for systems and CLI tools (Argus). Python for AI services and event-driven backends (Silicon Cortex, Meridian). TypeScript for full-stack and visualization (ArchByte). Kubernetes, Kafka, NeonDB, Redis, Docker, Cloudflare across projects. Claude, OpenAI, and Gemini for LLM integration.

Let's build something.

Production AI systems. Shipped, not pitched.

Schedule a CallArrow rightMore Showcases