AI lets product managers, designers, data scientists, and engineers move farther into one another's work, especially during ideation and prototyping. Stone argues that this fluidity does not remove the comparative advantage of each craft: product still frames the what, engineering owns scalable implementation, and data science judges whether evidence can be trusted.

Cross-functional teams can prototype before engineering is at the front of the line
Fluid roles help when the business problem is clear and partners remain aligned
Functional expertise still defines what good looks like
Great engineering, data science, and creativity remain scarce

“I still see a craft excellence that's really important in the disciplines that I don't think is going away anytime soon, even if there's fluidity…”
Elizabeth Stone

“I still find great engineering to be scarce, great data science to be scarce, great creativity to be scarce.”
Elizabeth Stone

#ai#product-teams#craft#roles

Myth Buster

Faster Code Does Not Make Deep Design Optional

Stone rejects the idea that faster coding should squeeze design expertise out of important product work. Designers can explore, iterate, and test more quickly with AI, but their deeper role remains making complexity invisible and preserving a coherent end-to-end consumer experience.

AI gives designers more options and faster iteration
Important priorities still deserve dedicated design work
Large consumer products can lose coherence when design is bypassed
The work changes, but the design mindset remains core

“But I think it would be a mistake to say design and deep design expertise and thinking get squeezed out just because we can write…”
Elizabeth Stone

“the product technology and design makes a lot of complexity invisible and makes for a seamless customer experience.”
Elizabeth Stone

#design#ai-tools#consumer-products#craft

Myth Buster

Why Netflix Is Still Hiring Junior Talent in the AI Era

Netflix continues to hire interns and new graduates because younger people bring openness to new workflows and current perspectives on entertainment and consumer behavior. Their development still requires craft mentorship, review, testing, diagnosis, and accountability, while learning can also flow from junior employees to experienced colleagues.

Junior hiring remains a critical part of Netflix's talent strategy
Younger people often arrive more native to new AI-enabled ways of working
Mentorship must continue to teach what excellent work looks like
Tool use does not remove responsibility for quality
Learning should run in both directions across generations

“We are still hiring junior people and they're really important to our talent strategy.”
Elizabeth Stone

“Probably the way we train and grow talent has to change because they're going to use different tools.”
Elizabeth Stone

#junior-talent#mentorship#ai-skills#craft

Hot Take· 3

Hot Take

Adaptable Generalists Are Rising as Narrow Specialization Shrinks

Netflix still needs rare specialists in areas such as encoding, playback, and production tooling, but Stone sees the broader talent mix shifting toward adaptable generalists. Specialized knowledge remains valuable when its owner is willing to question old tools, learn adjacent layers, and imagine the future version of the problem.

Netflix expects fewer narrow specialists than five or ten years ago
Engineers increasingly need to navigate multiple layers of the stack
Deep subject expertise remains useful in genuinely scarce domains
A specialist's willingness to learn and innovate determines whether depth stays valuable

“as a general rule, uh, compared to five or 10 years ago, I would believe we have fewer specialists and more people who are generalists…”
Elizabeth Stone

“I think the mindset now needs to be I can learn that quickly.”
Elizabeth Stone

#generalists#specialists#hiring#adaptability

Hot Take

Engineers May Write Less Syntax but Still Need System Fluency

Stone separates manually writing code in a particular language from understanding how code, systems, and products behave. Agents may generate more implementation, but humans still need enough fluency to judge whether a product is good, diagnose failures, understand unfamiliar output, and recover when systems break.

Writing syntax and understanding systems are different capabilities
Agent-generated code can be difficult to follow even when it performs better
Engineers still need to judge quality and expected behavior
Fast failure and recovery depend on understanding what was built
Tests and rationalization must evolve for unfamiliar AI-generated systems

“I think there's a difference between being able to write lines of code in a particular language like Python or C, and understanding how code,…”
Elizabeth Stone

“And I don't think the latter is going away.”
Elizabeth Stone

#engineering#coding-agents#systems#debugging

Hot Take

AI Can Amplify Storytelling, but Humanity Remains the Backbone

Stone can imagine AI playing a material role in production, but not compelling storytelling without humans at its center. Human creators understand connection, performers bring emotion to life, and technology is most valuable when it amplifies those qualities rather than trying to remove them.

Storytelling depends on understanding what connects with people
Human performance carries emotion that Stone finds difficult to replace
AI can materially shape how a production looks and feels
Technology should amplify human storytelling rather than erase its backbone

“I don't see the version of it that doesn't have the human as the backbone.”
Elizabeth Stone

“storytelling has been a key part of community and social networks and human feeling and connection.”
Elizabeth Stone

#storytelling#human-creativity#ai-content#entertainment

Explainer· 2

Explainer

Where AI Already Fits Across Content Production

Netflix applies AI beyond coding and product prototypes throughout the content lifecycle. Stone points to creative pre-visualization, post-production changes, localization, subtitles, dubbing, trailers, images, and artwork, while keeping filmmakers and creators in control of the vision.

Pre-visualization helps creators explore an idea before production begins
Post-production tools can relight, reframe, reshoot, or change dialogue
AI supports localization through subtitles and dubbing
Promotional assets can be created at scale to help titles find audiences
The same capabilities can support advertising and marketing

“Gen AI is a big step function in where the impact can be in creative ideation.”
Elizabeth Stone

“We've used them to think about how to create promotional assets at scale, how to localize in subtitles and dubs.”
Elizabeth Stone

#content-production#generative-ai#localization#post-production

Explainer

The Future of Entertainment Is a Personalized World, Not One Format

Netflix expects entertainment to span film, TV, live events, games, podcasts, devices, and different moments of the day. The product challenge is to turn that breadth into a seamless, personalized, immersive, and interactive journey instead of leaving members with a fragmented catalog that is hard to explore.

Consumer expectations now span formats, devices, and moments of the day
Netflix is expanding beyond traditional film and television
A member journey can connect podcasts, shows, and cloud games
Greater breadth raises the importance of discovery and engagement
Personalization must make a fragmented entertainment landscape feel coherent

“the future of entertainment isn't going to be one thing, and it's going to have to be more personalized, more immersive, more interactive”
Elizabeth Stone

“we've got to make discovery and engagement much easier than it feels today.”
Elizabeth Stone

#entertainment#personalization#multi-format#discovery

Story· 1

Story

Netflix's AI Story Started Long Before Generative AI

The Netflix Prize and years of machine-learning work gave the company a long head start on AI applications. Personalization remains the central problem: as Netflix expands beyond film and TV into games, live programming, and podcasts, matching the right title to the right person at the right moment becomes harder and more important.

The Netflix Prize showed the company's early commitment to machine learning
Personalization has long been central to the member experience
A broader catalog increases both the difficulty and value of discovery
AI is a tool for entertainment outcomes, even if Netflix is not branded as an AI company

“this is not new to us, that especially for personalization, it's been central to delivering a great experience to members.”
Elizabeth Stone

“You want to personalize right title for the right person at the right moment.”
Elizabeth Stone

#netflix-prize#personalization#machine-learning#discovery

Tool· 1

Tool

Turn Decades of Company Knowledge Into an AI Head Start

Netflix uses AI to distill past experiments, consumer research, metrics, and stakeholder knowledge into a faster starting point for new questions. The output is a head start rather than a final answer: people still form their own view and work with experts to confirm the data and interpretation.

AI can search and synthesize experiments and research that were previously hard to find
Faster retrieval accelerates hypothesis generation and analysis
Business functions can arrive with stronger initial hypotheses
Data scientists still validate sources, interpretation, and judgment

“AI is very powerful at distilling information, looking across a broad set of things, doing an analysis around it, getting to the core of here's…”
Elizabeth Stone

“I would hesitate to rely on that exclusively, but I think it's a head start.”
Elizabeth Stone

#knowledge-management#ai-analysis#experimentation#data

Takeaway· 3

Takeaway

Win AI Talent With the Problem, Not the Frontier-Lab Arms Race

Netflix competes for AI talent by being explicit about the kind of work only its domain can offer. People drawn to foundational model research may prefer frontier labs, while people energized by applying technology to global consumer products and entertainment can find a distinct mission at Netflix.

Different talent personas are motivated by different problem spaces
Netflix hires for passion about applied technology and entertainment
Consumer scale and global reach are part of the role's appeal
A company should articulate what its domain uniquely enables talented people to build

“You have to love entertainment, you have to love consumer products at scale, you have to love the global nature of that.”
Elizabeth Stone

“If instead you're inspired by some of the foundational work that the frontier model companies are doing, which is exciting in its own way, it's…”
Elizabeth Stone

#talent#recruiting#applied-ai#mission

Takeaway

Enable Every Creator Instead of Prescribing One AI Workflow

Netflix does not treat creators as having one acceptable relationship with AI. It supports filmmakers who reject AI, creators who want to explore new AI-enabled possibilities, and everyone between, with the goal of helping each creator bring their own vision to life.

Some creators see AI as incompatible with their process or vision
Others use it to explore stories, quality, and creative possibilities
Netflix supports the full spectrum rather than mandating one workflow
Flexible tools and partnerships follow from a creator-enablement posture
New formats can coexist with traditional film and television

“Netflix's role in this is to enable creators with whatever tools they want to use to bring their vision to life.”
Elizabeth Stone

“to really have a creator enablement view rather than a prescriptive, like we only do this one way.”
Elizabeth Stone

#creators#generative-ai#filmmaking#creative-tools

Takeaway

Do Not Lose the Consumer Product in the Technology

Stone closes by warning that an industry fascinated by technical capability can lose sight of the actual objective. The inspiring output is not the underlying AI itself, but a consumer product and entertainment experience that people around the world genuinely love.

Pure technology discussion can obscure the customer outcome
AI and product infrastructure are means rather than the final goal
Great consumer products and entertainment remain the measure of success
Builders should keep the human experience visible during rapid innovation

“we spend a lot of time sometimes talking about the the pure tech or the capability, and we sort of lose the forest for the…”
Elizabeth Stone

“We're trying to build great consumer products that people love.”
Elizabeth Stone

#consumer-products#product-strategy#ai#customer-value

Frameworks in this episode

AI Fluency Overlay

AI Paved Paths

Excellence as an Operating System

One-Click Zoom-Out

The Keeper Test

Insights & moments

⚡Myth Buster· 3

AI Blurs Product Roles Without Erasing Craft

Faster Code Does Not Make Deep Design Optional

Why Netflix Is Still Hiring Junior Talent in the AI Era

◆Hot Take· 3

Adaptable Generalists Are Rising as Narrow Specialization Shrinks

Engineers May Write Less Syntax but Still Need System Fluency

AI Can Amplify Storytelling, but Humanity Remains the Backbone

✶Explainer· 2

Where AI Already Fits Across Content Production

The Future of Entertainment Is a Personalized World, Not One Format

❝Story· 1

Netflix's AI Story Started Long Before Generative AI

✦Tool· 1

Turn Decades of Company Knowledge Into an AI Head Start

▲Takeaway· 3

Win AI Talent With the Problem, Not the Frontier-Lab Arms Race

Enable Every Creator Instead of Prescribing One AI Workflow

Do Not Lose the Consumer Product in the Technology

Myth Buster· 3

Hot Take· 3

Explainer· 2

Story· 1

Tool· 1

Takeaway· 3