Animation AI Studio

Portfolio Demo Platform

AI animation workflow studio for shot planning, generation jobs, monitoring, and review.

This project turns a research-heavy AI animation codebase into a demoable product surface: queued jobs, mock-safe sample data, result artifacts, health checks, and an interview-ready walkthrough that can run without private API keys or GPU hardware.

Start local demo GitHub repo
Web UI demo
4 seeded scenarios
100% demo-safe flow
SSE live progress
Shot Planning Completed
Live Monitoring Running
Failure Handling Documented
FastAPIbackend API
SQLitelocal job store
Vanilla JSstatic frontend
GitHub Pagesportfolio site

Screenshots And Demo Video

Public media assets for interview review.

These screenshots are captured from the runnable local demo mode. The video is a short reel of the portfolio page, seeded job dashboard, result browser, and system monitor.

Seeded Web UI job dashboard with completed, running, and failed demo scenarios
Job dashboard with seeded demo scenarios
Result browser showing demo output folders
Result browser for generated artifacts
System monitor showing CPU, RAM, GPU, disk, and usage chart
System monitor and realtime metrics
Portfolio landing page for Animation AI Studio
GitHub Pages portfolio landing page

Demo Scenarios

Built for screenshots and a 3-minute interview walkthrough.

Shot Planning

Shows project-to-shot thinking, manifests, prompt strategy, and reviewable production artifacts.

Screenshot: job details + summary artifact

Provider Routing

Demonstrates how image/video generation providers are abstracted behind workflow-ready jobs.

Screenshot: provider job + gallery output

Live Monitoring

Displays queue state, progress stages, logs, and CPU/RAM/GPU monitoring through FastAPI.

Recording: submit job, watch progress

Failure Handling

Makes external runtime gaps explicit, with clear error states instead of silent broken flows.

Screenshot: failed job + stderr log

Architecture

Local-first full-stack demo with clear extension points.

Frontend

Static Web UI with dashboards, job pages, result browser, and system monitor.

API

FastAPI service exposing health, jobs, stats, results, action, image, and monitoring endpoints.

Data

SQLite job database plus repo-local demo outputs under outputs/demo.

Pipeline

Demo mode produces deterministic artifacts; full mode can hand off to existing batch scripts.

Recording Flow

Recommended demo script

  1. Open the Web UI dashboard and point out seeded job states.
  2. Browse a completed scenario and show summary, gallery, and quality artifacts.
  3. Submit a demo-mode job and watch the status/progress update.
  4. Open system monitoring to explain the production-readiness surface.
  5. Close with architecture: static frontend, FastAPI, SQLite, repo-local artifacts.

Interview signal

The project is intentionally framed as an orchestration platform, not just a model script. It highlights product thinking, failure-mode handling, API design, data modeling, and deployable portfolio presentation.

Run Locally

Demo mode does not require private keys, model weights, or GPU access.

python scripts/demo/seed_demo_data.py --reset
WEB_UI_CONFIG_PATH=configs/web_ui/demo.yaml \
  python -m uvicorn web_ui.backend.main:app --host 127.0.0.1 --port 8000

Open http://127.0.0.1:8000. The static portfolio page is deployable through GitHub Pages from portfolio-web/; the interactive API demo should use Vercel, Render, Railway, or a VM because it needs a running FastAPI process.