Defining a Use Case
The use case sets the path
The story: Before a tailor cuts any fabric, they talk to the customer: what's the occasion, what's the budget, what do they like? They might ask a senior tailor for advice. Then they turn all that into exact measurements. A vague "make me something nice" leads to a suit nobody wanted.
In AI/AWS terms: Defining the use case names the problem, gathers requirements, and aligns stakeholders. Teams analyze the problem, consult subject matter experts, and translate business needs into technical specifications.
A business use case is a structured narrative of how a system should behave from a stakeholder's point of view. It starts with a brief description of its purpose.
For the exam: A business use case describes the system's behavior from the stakeholder's point of view, starting with its purpose.
Metrics for generative AI use cases
The story: How does the tailor know the new sewing machine was worth it? Wages spent per suit went down. Each suit takes fewer hours. The stitching is better. Customers are happier and recommend the shop. More suits get finished, with fewer mistakes.
In AI/AWS terms:
- Cost savings: lower labor costs and more efficient processes
- Time savings: less time per task
- Quality improvement: accuracy, coherence, and creativity of the output
- Customer satisfaction: satisfaction scores, net promoter score (NPS), sentiment
- Productivity gains: output volume, error rates, task completion time
For the exam: NPS and satisfaction scores measure customer satisfaction. Output volume and error rates measure productivity.
Ways generative AI addresses business problems
The story: The machine can do the boring hemming by itself, suggest which fabric suits a customer, make each suit fit its owner exactly, design new patterns for the shop window, and help the tailor try out styles nobody has tried.
In AI/AWS terms:
- Process automation of repetitive tasks
- Augmented decision-making with insights and recommendations
- Personalization and customization for customers
- Creative content generation for marketing, entertainment, and education
- Exploratory analysis and innovation to find new ideas
For the exam: Generative AI helps through automation, better decisions, personalization, creative content, and innovation.