Applied AI Lead

The Role
Over 100 people at Fair Trade USA use Claude Enterprise, with 80 percent adoption and 6.5 agentic actions per prompt, saving 377 hours a month. Onboarding is solved. Depth is not. The Applied AI Lead owns depth: finding, testing, and proving the ways AI changes how the organization's work gets done, function by function, and measuring what changed.

Adoption is wide and shallow. Most use is generic, and the gains still available are in function-specific workflow patterns that nobody owns. The technical hires in Applied Intelligence build systems. You carry those systems into daily work, test what holds, and report what did not.

You run experiments with real autonomy: you set the experiment calendar inside quarterly priorities and report monthly. You also back up administration of Claude Enterprise, so the AI program does not depend on any one person.

Key Responsibilities

Workflow Experiments

• Select a function, design a workflow pattern with the people who do the work, and test it.
• Measure what was displaced and what merely got faster. Publish the result, scale what holds, and retire what does not.
• Set the experiment calendar inside quarterly priorities and report monthly.
• Document successful patterns and lessons so other teams can reuse them.
• Track emerging AI capabilities and test the ones that matter to a function's actual work.

Patterns & Integrations

• Build prompt, agent, and integration patterns each team needs, reusable rather than personal.
• Build with prompts, agent workflows, light integrations, and basic scripting.
• Work with Applied Intelligence engineers to carry the systems they build into daily work, and bring back what users find.
• Refine patterns based on user feedback and measured outcomes.

Measurement

• Own enablement measurement. Only displaced work is worth reporting.
• Define the metric and the baseline for each experiment before it starts.
• Analyze usage data to find where use is shallow and where depth is paying off.
• Report monthly to leadership, including what did not work. A measurement function that only produces good news is worthless.

Governance & Enablement

• Back up Claude Enterprise administration, governance and usage standards, and onboarding patterns, so the AI program does not depend on any one person.
• Maintain AI governance practices and usage standards.
• Run the all-staff AI series.
• Keep documentation current enough that another person could run the AI program.

Cross-Function Collaboration

• Work directly with the people who do the work in each function, and start from their operational problems.
• Advise leaders and teams on where AI changes the work and where it does not.
• Explain findings, recommendations, and results clearly to technical and non-technical audiences.
• Take a no as a prompt to return a second time with better evidence.

Key Qualifications / Experience

• You have changed how a team works, not only trained it, and can show the before and after.
• You can build what you design: prompts, agent workflows, light integrations, and basic scripting. Not an engineer, not a trainer; a builder of working patterns.
• You have credibility across functions. People outside your team already come to you.
• Hands-on experience with AI platforms, large language models, and workflow automation tools.
• Strong analytical skills: you measure outcomes, separate displaced work from faster work, and report results plainly.
• You are comfortable being told no and going back a second time, and willing to report that something did not work.
• Experience designing and leading cross-functional initiatives that deliver measurable outcomes.
• You work well in experimentation, ambiguity, and continuous learning.
• 1-3 years of experience in product management, business transformation, process improvement, AI enablement, or a related field.