AI-to-NLE Integration Suite
Comprehensive Model Context Protocol server for professional editing software. Modular architecture spanning editorial, color, compositing, media management, and delivery — the most complete integration of its kind.
I build AI systems for film and broadcast.
30+ years in post-production finishing across every major studio. Now designing locally-hosted AI pipelines, multi-agent orchestration, and creative tools for the entertainment industry.
Selected Work
What I Build
AI infrastructure, creative tools, and production workflows — from concept through deployment.
Multi-Agent Orchestration Framework
Purpose-built agent system for entertainment workflows. Specialized agents organized into operational pods, backed by relational database architecture with secure tunneled access.
On-Prem Inference Architecture
Three-tier, air-gapped-capable inference stack designed for studios with strict content security requirements. No cloud dependencies. No vendor lock-in.
Multilingual Localization Pipeline
End-to-end subtitle and dubbing pipeline for international film distribution. Automated translation, QC, and territory-specific delivery packaging across dozens of languages.
John Lennon Educational Tour Bus — AI Platform
Real-time AI platform for the John Lennon Educational Tour Bus, deployed at NAMM 2026 and running permanently. Vision, transcription, generative chat, and audience interaction — all on local GPU inference. Full case study below.
Read case study ↓AI-Assisted Music Production
Active recording artist across multiple aliases on all major platforms. Exploring generative audio workflows including custom model training on proprietary music catalogs.
Case Study
John Lennon Educational Tour Bus — AI Platform
Designed, built, and deployed a real-time AI experience for one of music education's most recognized programs — running permanently on local GPU hardware.
The Challenge
The Lennon Bus needed a live AI demonstration for NAMM 2026 — the largest music industry trade show in the world — that could handle thousands of concurrent interactions with zero cloud dependencies, running entirely on an NVIDIA DGX Spark contributed to the program days before the show.
Inference Stack
- NVIDIA DGX Spark (local cluster)
- Nemotron Vision 12B — image understanding
- Nemotron Chat 30B — conversational AI
- Parakeet TDT — real-time speech-to-text
- BROBOT Jungle — custom generative model
Interaction Layer
- Twilio SMS (toll-free inbound)
- Telegram bot
- FastAPI backend
- Live photo gallery with AI-generated captions
- Real-time status dashboard
The Approach
Attendees texted a toll-free number or messaged a Telegram bot.
Inbound media and questions were routed to local models
for vision analysis, captioning, and conversational response —
all processed on-site with sub-second latency.
A live web gallery updated in real-time as contributions came in,
building a collaborative record of the event.
The Outcome
Ran the full duration of NAMM 2026 with zero downtime.
Hundreds of interactions processed across SMS and Telegram.
Demonstrated to NVIDIA executives on-site.
Resulted in permanent AI Fellow appointment
and an ongoing deployment now running year-round.
Background
Where I Come From
Three decades inside the machine — finishing rooms, edit bays, and festival stages. Now rebuilding it all with AI.
Post-Production Finishing
Online editing, color, delivery, and pipeline design for film
and episodic television across every major studio.
AI Systems Architect
Designing and building locally-hosted, security-compliant AI infrastructure
for entertainment and broadcast post-production.
AI Fellow
Designing and deploying real-time AI demonstrations
for a nationally touring education program.
Filmmaker & Musician
Colorist, cinematographer, composer, sound engineer, recording artist,
and vinyl DJ with international festival credits. Recent color credit: American Hemp Farmer (PBS).
Contact
Let's Talk
Open to roles where AI infrastructure, post-production pipelines, and creative technology intersect. Let's connect.