(2:19 Q&A Video) “Improve AI Learning With Prompts, Documentation, and Practice”

In this 2:19 video, Alysha McCree shares how her team learns to use AI through ongoing experimentation and peer teaching. They build documentation and prompt libraries to help one another get better results. She connects those internal practices to how the company helps software engineers work more efficiently. Her approach is to experiment, document, standardize, and continue optimizing what works.

AI capability is changing quickly, but teams don’t become skilled simply by getting access to new tools. Alysha McCree, Head of Marketing at FPT Americas, describes a learning process her team uses as models and methods continue to evolve. People experiment individually, share what they learn, teach one another, and build documentation and prompt libraries. That routine matters for marketing leaders because it turns scattered discoveries into shared practice.

Alysha also connects the same habits to software engineering work that can help clients go to market faster and reach value sooner. Her example offers a practical way to develop AI fluency while creating repeatable improvements across a business.

“What we do is individually experiment, and we learn, and then we come together, and we teach each other, and we try to build documentation, prompt libraries of our own that we find really work well in order to help each other leverage AI tools better or do better prompting. Experiment, document, standardize, and then continually be in a position of optimizing that is the best that you can do.”– Alysha McCree

To learn more, watch the video or read the article below.

To catch the full interview with Alysha at “Unlocking the Business Value of AI Personalization.” CLICK HERE.

Article: "Improve AI Learning With Prompts, Documentation, and Practice"

This article is based on an interview with Alysha McCree, Head of Marketing at FPT Americas

Build AI Skills Through Experimentation and Peer Teaching

A team’s AI learning should continue even after initial training. Alysha says her team spends “a lot of time talking about AI and learning from one another” because models and tools keep changing. Experimentation gives people a way to discover what works in their actual tasks instead of assuming that general familiarity equals skill. Team leaders can make this practical by giving employees room to test approaches, then bringing useful findings into shared conversations. This keeps learning connected to real work and helps teams adapt as capabilities change. Managers can protect a small amount of time for trials and make it easy to share useful examples. A regular cadence gives learning momentum without requiring a large formal program.

The peer-learning step makes individual discoveries more useful. Alysha explains, “What we do is individually experiment, and we learn, and then we come together, and we teach each other.” That sequence spreads practical knowledge across a group instead of leaving it with the person who first found a helpful technique. A team can use brief show-and-tells, examples of effective prompts, and discussion of where an output needed human review. The purpose is to build confidence through use and help colleagues learn from one another’s experience.

Turn Useful Discoveries Into Shared Documentation

Experimentation alone can produce uneven results if each employee has to rediscover the same lessons. Alysha’s team builds documentation and prompt libraries to capture approaches that work well. These shared resources can help people ask clearer questions, provide better context, and use AI tools more consistently. The library does not need to be elaborate to be useful. Teams can record the task, prompt, relevant inputs, review steps, and the conditions where the approach is effective, then update the guidance as the tools evolve.

Alysha describes the goal as creating “documentation, prompt libraries of our own that we find really work well.” This keeps the emphasis on tested team knowledge rather than generic collections of prompts. Alysha says the library helps colleagues “leverage AI tools better or do better prompting.” Leaders can encourage employees to explain why a prompt worked and what to check before using the result. Keeping examples tied to specific tasks makes the resource easier to search and maintain.

That makes documentation more transferable and reduces the risk of treating a useful example as universally reliable. It also gives new team members a clearer starting point and helps experienced employees build on prior work. Over time, a curated library can show where a method is reliable, where it needs more context, and what a person should check before using the output.

Standardize What Works and Keep Improving It

Once a team has learned and documented an effective approach, it can decide what should become a shared standard. Alysha initially describes standardizing, documenting, and experimenting, then revises the order to “experiment, document, standardize” and continue optimizing. That order reflects an important operating choice: first learn what is useful, then capture it, and only then formalize it. A process should not become standard just because it is easy to write down. It should be supported by experience and remain open to adjustment. Alysha recommends “continually being in a position of optimizing” the practice.

Alysha connects this method to how FPT augments software engineers so they can develop code faster and move through the software development life cycle more efficiently. Those improvements can help clients go to market faster or speed their time to value. Her example shows how internal learning can contribute to customer outcomes when leaders connect team practices with delivery work. Teams can track whether a shared AI practice improves consistency, turnaround time, or the ability to complete valuable work, then refine it as results and tools change.

Conclusion

The team-learning model Alysha McCree describes is straightforward: let people experiment, bring their discoveries together, document what proves useful, standardize the repeatable methods, and continue optimizing. Each step solves a different problem. Experimentation helps teams adapt, peer teaching spreads knowledge, documentation preserves it, and standards make it easier to apply consistently. Ongoing review keeps those practices responsive as AI capabilities and business needs change.

For leaders, AI learning is measured not by how many prompts a team collects. It is whether employees can use shared knowledge to improve important work and support better customer outcomes. By treating learning as an operating practice, organizations can make AI capability more durable and connect it to faster delivery and time to value. The habit gives teams a way to improve as their work and tools continue to change.

ELEVATING CONTENT MARKETING STANDARDS RAISES A CRUCIAL QUESTION:
Why shouldn’t the very act of creating thought leadership content also spark new conversations with in-pipeline deals, top prospects, and high-priority customers for revenue expansion? Or serve as the cornerstone for learning and applying voice-of-customer strategies? That’s the power of ABM Podcasting—engaging B2B buyers and customers in meaningful conversations that uncover essential go-to-market insights, support long sales cycles, reinforce your market perspective, build peer-level trust, and dismantle objections that kill deals.

Steven MacDonald

Steven MacDonald is the founder of Content Strategies, CEO of a MarTech SaaS Company, fractional CMO consulting with leading B2B companies and former Director of Strategy and Client Service at top ten digital marketing agencies.

Follow Steve on LinkedIn.

Steven MacDonald

Follow Steve on LinkedIn.