Instructions:

Choose one scenario, them anwser the questions and share your anwsers in your 1.1 meeting with your EM

1. Scenario 1:

   • Task: Your company is launching an e-commerce website and needs a model to generate short product descriptions for hundreds of items.

   • Questions to Answer:

     • Is this task simple or complex?

     • Which AI model (e.g., GPT-3, GPT-4) would you choose and why?

     • Do you think using a more advanced model (GPT-4) is necessary for this task, or would a simpler model work just as well?

     • What are the potential cost implications of using a more advanced model, and is it justified by the task’s needs?

     • What security and compliance considerations would you need to take into account for handling customer data (if applicable)?
       
       

2. Scenario 2:

   • Task: You are developing a customer service chatbot for a healthcare organization. The chatbot needs to understand complex queries about patient treatment plans, health conditions, and legal matters related to patient privacy.

   • Questions to Answer:

     • Is this task simple or complex? Justify your answer.

     • Given the complexity of the task, which AI model would you choose (GPT-3, GPT-4, or a specialized model)?

     • Assess the performance metrics (e.g., accuracy, speed, precision) needed for this use case. How would you measure success?

     • What security and compliance concerns would you need to address in the context of healthcare data (e.g., HIPAA)?

     • How would you ensure the chatbot remains scalable as more users interact with it?
       
       

3. Scenario 3:

   • Task: Your team is tasked with creating a custom AI model for a financial institution to detect fraudulent transactions in real-time. The model should analyze large amounts of data, including transaction history, user behavior, and market trends.

   • Questions to Answer:

     • Is this task simple or complex? Justify your answer.

     • What decision criteria (cost, context, performance metrics, scalability) would you consider when selecting an AI model for this task?

     • Which model would you choose for this task, and why (GPT-3, GPT-4, or a specialized financial model)?

     • How would you avoid over-engineering the model while ensuring its effectiveness?

     • What security and compliance issues should you consider when handling sensitive financial data (e.g., GDPR, financial regulations)?
       
       

4. Scenario 4:

Task: Your company wants to create a content generator to produce social media posts, blogs, and product advertisements for a consumer goods brand. The brand's tone is friendly and engaging, and the content should appeal to a broad audience.

4. • Questions to Answer:

     • Is this task simple or complex? Justify your answer.

     • Which model would you choose for this task, and why (GPT-3, GPT-4, or another model)?

     • How would you assess the balance between cost and performance? What performance metrics would you consider?

     • What considerations around scalability should you account for if the content generation process needs to handle multiple languages or a high volume of posts?

     • How would you ensure the generated content remains aligned with the brand’s voice and meets security and compliance standards?