Generative AI for Story Development
Generative AI at Netflix, turning scripts into visuals, insights, and shared understanding.
An AI-first storytelling platform that turns scripts into generative visuals, insights, and collaborative narratives, helping teams understand stories and make decisions faster.
My contributions
- Led design end to end for Story Lab, owning vision, interaction model, and quality bar from zero to launch.
- Set the long-term AI product direction while unblocking the team's immediate roadmap week over week.
- Partnered directly with cross-functional leaders across 8 orgs, product, engineering, research, and studio, to align on one source of truth.
- Coached and mentored designers on the team, raising the bar for craft, systems thinking, and how we work with AI.
- Drove executive reviews and vision work that turned Story Lab into a company-level bet with sustained investment.
- Role
- Lead Interaction Designer, AI
- Client
- Netflix
- Year
- 2025-Present
- Discipline
- AI, Product
- Vision
- AI interaction model
- Collaborative UX
- Design system
The problem
Every show at Netflix starts as a script, but for years that script lived as a flat PDF. Writers, producers, executives, and marketing teams each had their own copy, their own notes, and their own interpretation. Coverage took weeks, insights were trapped in email threads, and eight different organizations were all reading the same story in eight different ways. The source of truth was scattered, and the cost of misalignment was real.
The approach
The industry was sprinting toward AI that writes. Our research kept pointing somewhere else. Writers were not asking for a machine to replace them; they were asking for help making sense of what they had already created. I reframed the brief from AI that writes to AI that helps people understand stories. The goal became clear: build tools that visualize, explore, and clarify creative intent so teams can communicate ideas as clearly as they understand them.
Explorations and iterations
Early prototypes were rough and fast: paper sketches, FigJam diagrams, and vibe-coded demos that let us test what teams actually needed before committing to polish. Each artifact captured a different question: how might chat work across scripts, where does Story Lab plug into the journey, what density feels right, and which directions are worth setting aside. We gave people control to shape, refine, and steer every result.
What we shipped
Story Lab
The creative vision for where Script Hub goes next: an AI-powered workspace where teams shape, refine, and visualize the story as characters, scenes, moodboards, and motion.
Story Lab, on the go.
The whole story, in your pocket.
Read the script, ask the hard questions, and shape the visuals from anywhere. The same source-grounded workspace, sized for the ten minutes between meetings.
A closer look
Tap through the moments that make Story Lab feel less like a tool and more like a second brain on set.
One workspace, whole story
Everything your story knows, in one place.
Drop in scripts, treatments, shot lists, budgets and lookbooks. Story Lab indexes it all, so every answer starts from your pages, not the internet.
Script Hub
The production platform used across Netflix to read, search, and analyze scripts. A family of AI-native capabilities that extend the script itself, every answer citing the line it came from.
Scripts widget: a consistent script experience across the Netflix ecosystem
Alongside Script Hub and Story Lab, I designed a Scripts widget that could be embedded across multiple Netflix apps. The goal was simple: no matter where a user is in the ecosystem, they should have fast, consistent access to the script and a reliable reading experience. Depending on the host app and context, the widget shows up in four forms.
Theme · 01
A · Inline Card
A self-contained card you drop into any page, a title dashboard, a production tracker, a feed.
Theme · 02
B · Launch Pill → Popover
A compact pill that lives in any app's toolbar.
Theme · 03
C · Full Modal Reader
The focused experience, a centered overlay over a dimmed host.
Theme · 04
D · Embedded Listen Row
A rich list row for feeds and libraries, the script as a playable, expandable item.
Where it goes next
Alongside what shipped, we mocked a set of future concepts to pressure-test the vision and give leadership something tangible to react to. Each one imagines a different way Story Lab could deepen understanding of the script, from structural analysis to production breakdowns to visualizing key scenes. Concepts, not commitments, but concrete enough to argue about.
What I learned
In AI products, trust is the feature
Everything else is decoration. The moment a user cannot trace an answer back to something real, they stop believing the tool. Grounding every summary, visual, and suggestion in the script itself turned skepticism into curiosity, and curiosity into daily use.
AI accelerates judgment, it does not replace it
The moments that matter most in storytelling are moments of judgment. AI's job is not to make those calls, it is to get people to them faster, with more context and less friction.
Design for the whole studio, not one role
Script Hub and Story Lab worked because they gave eight different organizations a shared surface. When the tool respects each team's craft, alignment stops being a meeting and starts being a byproduct.