Director of AI: The Job Title You Didn’t Know You Already Have

The panic-inducing headlines about AI stealing jobs have begun to fade, replaced by a more nuanced reality: AI isn’t taking your job — it’s becoming your most productive team member. The real challenge for organizations isn’t whether to adopt generative AI, but how to orchestrate a workforce where human expertise and machine capabilities combine to create something greater than the sum of their parts.

“GenAI hasn’t replaced expertise on our curriculum team, it’s just changed the way we apply our expertise to the daily job,” explains Kiki Carter, senior staff curriculum developer at Cockroach Labs. Carter has been leading the company’s integration of AI into their curriculum development processes, achieving what once seemed impossible: a 5-time acceleration in productivity while simultaneously raising quality standards.

The director’s chair

What’s emerging from forward-thinking companies like Cockroach Labs is a new organizational paradigm where employees aren’t being replaced by AI — they’re being elevated to directors of AI resources. This shift demands a fundamental rethinking of what constitutes “expertise” in the AI-augmented workplace.

“Our curriculum developers are becoming more like technical directors who orchestrate resources, including AI, to create high-quality learning experiences,” says Carter. “The technical depth is still crucial — you can’t direct what you don’t understand — but the expression of that expertise has evolved.”

This director-level mindset requires employees to develop what might be called “AI literacy” — not just the ability to use AI tools, but the sophistication to guide them effectively. Carter describes it as similar to managing junior team members: "You need to clearly communicate what you need and verify the output.”

For CHROs and talent acquisition leaders, this represents a seismic shift in hiring criteria. The most valuable employees won’t necessarily be those who can execute tasks the fastest, but those who can effectively orchestrate AI capabilities while applying their domain expertise at strategic inflection points in the workflow.

The quality-speed paradox, resolved

Perhaps the most striking revelation from Cockroach Labs’ experience is how AI has resolved what previously seemed like an irreconcilable tension between speed and quality.

“Speed of development has been a great outcome, but it’s not the speed alone that we’re excited about,” Carter notes. “We are equally pleased with how much time we’ve reclaimed for expert-level refinements and research that our team does. And that directly impacts the quality of the content we build.”

This represents a profound shift in how organizations should think about productivity gains from AI implementation. The goal isn’t simply to do the same work faster, but to reallocate human capital toward higher-value activities that machines can’t perform.

In practical terms, this means that AI acceleration isn’t just about efficiency metrics — it's about creating capacity for excellence. At Cockroach Labs, curriculum developers now spend less time on initial drafting and more time on deep refinement, resulting in learning materials that are both produced faster and of higher quality than before.

The new talent stack

For CHROs, Carter’s insights offer a blueprint for the skills taxonomy needed in an AI-integrated workplace. Technical depth remains essential — employees still need domain expertise to identify when AI produces incorrect information or misses critical nuances. But layered on top of this foundation are emerging competencies:

1. Prompt engineering: The ability to provide AI with precise context and craft queries that yield optimal results
2. Output verification: Critical thinking skills to evaluate AI-generated content for accuracy and quality
3. Process orchestration: Strategic thinking about when and how to deploy AI within complex workflows

“When hiring, we continue to look for technically strong candidates who have experience writing code, working with databases, and explaining complex concepts,” says Carter. “But they also need to be comfortable with AI as a collaborative tool and have well-honed discernment and the ability to pick up where AI leaves off if needed.”

This points to a future where human-AI collaboration becomes a core competency rather than a specialized skill set. Organizations that fail to develop this capability across their workforce risk falling behind competitors who effectively leverage these new collaborative dynamics.

Stress-testing the human-AI system

Perhaps the most overlooked aspect of AI integration is resilience planning. As organizations become increasingly dependent on AI systems, how do they ensure continuity when those systems fail or produce unexpected results?

Carter draws fascinating parallels between Cockroach Labs’ approach to database resilience and their AI content pipeline: “Just as the database is designed to survive node failures, our content pipeline is designed with multiple validation checkpoints.”

This distributed systems thinking applied to workforce planning represents a new frontier for organizational resilience. Rather than treating AI as a monolithic solution, forward-thinking organizations are implementing what Carter calls “content circuit breakers” and recovery protocols similar to transaction logs.

“It’s really about applying the same distributed systems thinking that makes CockroachDB resilient to our content creation process — designing for failure rather than hoping it won’t happen,” she explains.

For CHROs, this suggests the need for workforce contingency planning that accounts for AI dependencies. What happens when the AI system hallucinates or encounters domain-specific edge cases? How quickly can human experts step in to maintain continuity?

Measuring the new collaboration

As organizations deepen their AI integration, traditional performance metrics become insufficient. Carter’s team is pioneering new approaches to measuring human-AI collaboration effectiveness, focusing not just on output metrics but on the quality of the interaction itself.

“We’re tracking quantitative measures like the volume of edits required as well as qualitative dimensions - categorizing edits by type, such as technical correctness, style adjustments, or detail enhancement,” Carter explains.

These emerging metrics point toward a future where organizations will need sophisticated frameworks for evaluating not just individual performance, but the effectiveness of human-AI collaborative systems. The goal, as Carter puts it, “isn't just to measure AI performance in isolation, but to optimize the entire collaborative workflow.”

The road ahead

What emerges from Cockroach Labs’ experience is a vision of work that’s neither dystopian nor utopian, but pragmatically transformative. AI isn’t eliminating the need for human expertise — it’s changing how that expertise manifests in daily work.

The organizations that will thrive in this new paradigm aren’t those that view AI as a cost-cutting measure to reduce headcount, but those that leverage it to elevate their workforce to higher-order tasks that drive innovation and quality.

As Carter's experience demonstrates, this is happening now, with real teams achieving productivity gains that were previously unimaginable without sacrificing quality.

CHROs and technology leaders need to start building the organizational capabilities and talent frameworks that will enable their teams to thrive in this new collaborative future. Or competitors will. 

Image credit: iStockphoto/Geerati