Portfolio thesis
Models become products when workflows make them usable.
I am a filmmaker and creative technologist exploring how AI workflows can turn generative video, 3D scene tools, evaluation systems, and inference infrastructure into repeatable production pipelines for artists, studios, brands, and enterprise teams.
Creative production anchor.
The portfolio should be read through the lens of S4: a feature documentary I co-directed, supported by practical AI/post workflows and a JRE clip that reached 2M-plus views.
S4 gives the work an end-to-end creative anchor: story, edit, collaboration, and audience.
Daily DaVinci Resolve practice connects AI experiments to actual editorial and finishing workflows.
ML pipelines such as de-aging, green screen removal, and neural rendering are positioned as production tools.
The repos show I can go deeper when a creative workflow needs custom evaluation, rendering, or deployment support.
The 60-second pitch.
Creative AI is no longer just prompt-to-video. The useful work is designing workflows that help teams control, evaluate, iterate, and deploy model outputs inside real production environments.
Four customer-facing workflows.
Each case study starts from a creative or enterprise problem, then maps back to technical exploration in the supporting repos.
Cinematic Control
Director-friendly control over camera, motion, force, style, and continuity.
023D Scene Review
Spatial review before animation, previs, or generated video handoff.
03Creative QA
Plain-language model evaluation for generated video review.
04Enterprise Rollout
Latency, routing, observability, and reliability for creative AI systems.
Why this fits creative workflow architecture.
The role is not only about building models. It is about pushing creative tools, explaining workflows to customers, training teams, translating feedback into product insight, and packaging experiments into reusable systems.
The work is framed around creative outcomes, not only benchmarks.
Every workflow is grounded in a technical exploration.
Adoption, training, reliability, and team workflows are explicit.
The portfolio values the path to the result, not just the result.
Technical exploration underneath.
The showcase leads with workflow value, then lets the repos show technical depth across video, 3D, evaluation, and infrastructure.
| Repo | Portfolio role | What it demonstrates |
|---|---|---|
| splat-web-viewer | Visual demo anchor | Interactive 3D review, SPZ assets, and browser-based spatial media tooling. |
| cinematic-controlnet | Creative control anchor | Shot control, cinematic inputs, and physical motion conditioning. |
| video-gen-eval | Evaluation anchor | Creative QA, model comparison, and failure explanation. |
| cloud-inference-platform | Enterprise anchor | SLO-aware deployment, latency control, observability, and routing. |