AI Fluency Overlay
Set one adaptable AI expectation across roles instead of freezing it into ladders
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 94%
The AI Fluency Overlay adds a shared expectation across the workforce without rewriting every career ladder around today's tools. The aspiration applies to all roles and seniority levels, but its evidence changes by function and career stage. Fluency can include an experimentation mindset, knowing where AI is useful or unreliable, building with it when relevant, and remaining accountable for the result. Leaders translate the overlay into realistic examples for design, engineering, data science, product, and business roles, then revisit those examples frequently as the technology evolves. Hiring should test the same qualities in the environment candidates will actually use, including access to AI tools. This keeps the organization adaptable while preserving craft standards and outcome-based judgment.
Origin
Netflix chose not to specify exactly how AI changes every level of each career ladder. Instead, the company placed an aspiration for AI fluency across all talent, then adapted its meaning by function, career stage, and changing technology.
Core principles
- 01AI fluency applies to every function and level
- 02Fluency means judgment as well as tool use
- 03Expectations should adapt faster than formal career ladders
- 04Technology should be used where it improves outcomes, not for its own sake
- 05Curiosity and comfort with change are non-negotiable
How to run it
- 1
Set the shared aspiration
State that AI fluency is expected across functions and levels. Keep the top-level expectation durable enough to survive tool changes.
Pro tip Frame fluency around better work and judgment, not mandatory use of a named product.
Watch out Do not equate fluency with using AI for every task.
- 2
Translate by role and stage
Define what useful fluency looks like for each function and career stage. A senior leader, designer, engineer, and new graduate can meet the aspiration in different ways.
Watch out A single technical benchmark will mismeasure non-engineering roles.
- 3
Assess judgment and experimentation
Look for comfort exploring new methods, evidence of hands-on use where relevant, and the ability to identify where AI should or should not be trusted.
Pro tip Ask candidates and employees what they use, what changed in their workflow, and how they validated the result.
- 4
Make evaluation realistic
Allow the AI tools people will use in the actual job during appropriate hiring and development exercises. Evaluate the quality of their process and output.
Pro tip Keep accountability for the final work explicit even when a tool produced part of it.
- 5
Refresh the evidence
Review role-specific examples frequently as capabilities and workflows change. Preserve the aspiration while updating what demonstrates it.
Pro tip Use recent work examples to update the overlay instead of waiting for a full ladder rewrite.
Watch out Static criteria can become obsolete within months during rapid technology shifts.
In the wild
Netflix allows candidates to use AI tools in coding interviews because those tools are part of the work. The evaluation can therefore focus on how candidates think, explore, validate, and remain responsible for the result rather than testing an artificial tool-free workflow.
→ Hiring evidence better reflects the environment and judgment the role now requires.
Netflix's senior leaders are also expected to develop deep AI fluency even when writing code is not part of their day jobs. Their evidence of fluency can center on judgment, applications, organizational change, and informed experimentation.
→ AI adaptation becomes a leadership expectation rather than a task delegated only to technical teams.
Common mistakes
Hard-coding today's tools
Embedding a current product or technique into every career level makes the ladder stale as capabilities change.
Rewarding AI use for its own sake
Fluency requires knowing when AI is useful and when review or a different method will produce a better outcome.
Exempting senior leaders
Treating AI as an execution-layer concern leaves decision-makers unable to guide changing work and talent expectations.
Is it for you?
Best for
Leaders updating hiring, development, and performance expectations during rapid changes in AI-enabled work.
Not ideal for
Organizations that have not yet identified any relevant AI-assisted work or cannot safely permit experimentation.
From the episode
Netflix CPTO on AI and the future of product and tech roles
Elizabeth Stone