My AI Agent Isn't Allowed to Decide Anything

Description and use

Shot‑Delivery Guardian uses AI to analyze post‑production stages and give explainable push‑back recommendations without affecting client‑approved or director‑flagged work, achieving 60 % accuracy in real tests.

Problem

A film production pipeline can miss deadlines if a step backs up, requiring a fast decision on which work to delay without affecting client‑approved or director‑flagged items.

Outcome

In 5 real tests, 3 matched the fixed rule’s choice, 0 broken rules, average 44 s response time, 60 % accuracy, demonstrating reliable, explainable AI support without autonomous decision‑making.

Workflow

• 1. Intake, rendering, color, quality checks, delivery, scheduling, test generation, and AI modules run independently.
• 2. Each module reports status to a shared connector.
• 3. AI queries three monitoring tools: overall backlog, work‑level metadata, and stage‑level timing.
• 4. AI applies rule to filter eligible shots and sorts by slack.
• 5. AI writes recommendation note to dashboard.
• 6. Human on‑call reviews note and makes final decision.

Evidence

Three out of five. That’s how often Shot‑Delivery Guardian, the system I built for Google’s Agentic Cinema hackathon, picks the right answer when it’s the one making that call, in real tests against the actual running system, not just on my laptop.

Source

Zero broken rules. It never once picked a piece of work the client had already signed off on. It never once picked something other work was waiting on. It never once touched anything the director had flagged as important.

Source

The AI runs this rule and reports the answer back, word for word. It doesn’t get to argue with it or dress it up to sound more sure of itself than it is.

Source