PaulDuvall

Objective:

Generate a structured case study in multiple formats (HTML, PDF, and others) from a 30-minute recorded conversation using the latest version of Claude via Amazon Bedrock.

Requirements:

1. Transcription: Convert recorded audio into text using Amazon Transcribe.
2. Case Study Generation: Process the transcription through Claude (latest version) via Amazon Bedrock to generate a structured case study based on the predefined Case Study format.
3. Output Formats: Export the case study in multiple formats, including HTML, PDF, and other relevant formats.
4. Efficiency & Scalability: The solution must follow best practices in terms of performance, error handling, and scalability.

Deliverables:

PaulDuvall

Objective:

Develop a structured list of targeted questions for recorded interviews with key project stakeholders—including engagement managers, consultants, and engineers—who directly contributed to delivering AWS-based solutions. The goal is to extract precise, actionable insights to facilitate the automated generation of compelling customer case studies.

Context & Analysis:

A review of AWS and Stelligent case studies reveals distinct archetypes, structural formats, and storytelling approaches. Understanding these patterns will ensure that the questions elicit the right level of detail and depth for compelling narratives.

Case Study Archetypes:

1. Transformation Stories – Focus on major shifts in efficiency, scalability, or innovation.
   • Example: 3M Health Information Systems automated pipeline provisioning using AWS Service Catalog, reducing setup time from days to minutes.

PaulDuvall

AWS Case Study Generator Prompt

You are an expert AWS Solutions Architect. You will help generate case studies based on audio transcripts. Your goal is to create detailed, technical case studies that showcase AWS best practices and real-world implementations.

Guidelines for Case Study Generation

When generating case studies:

1. Focus on technical details and AWS services
   • Specify exact services used (e.g., Amazon ECS, AWS Lambda)

• Include configuration details where relevant

PaulDuvall

AWS Case Study Analysis Guide

Objective:

Analyze the case studies from the provided sources to identify common archetypes, structural formats, and storytelling techniques used in their presentation. The goal is to extract key patterns and variations that can inform the creation of compelling, structured case studies.

---

Analysis Criteria:

1. Archetype Identification

PaulDuvall

"""
Case Study Generator - Automated tool for generating case studies from audio transcripts
using AWS services and AI.
"""

from __future__ import annotations

import json
import mimetypes
import os

PaulDuvall

generative AI Prompts: AWS Certified AI Practitioner

Prompt 1: Multiple-Choice Only

Base all questions on the https://d1.awsstatic.com/training-and-certification/docs-ai-practitioner/AWS-Certified-AI-Practitioner_Exam-Guide.pdf and derive from the attached content (Mapping of AI, ML, Deep Learning, and Related Concepts.pdf, practice-exam.pdf, terms.png). Create concise case study questions, each presented in a short paragraph outlining a realistic challenge. The questions should ask for the best option(s) based on specific criteria and require critical evaluation. Each case study will be followed by multiple-choice options. Please generate a mix of medium-complexity questions, blending AWS AI/ML services with general AI/ML concepts.

Question Format

1. All questions should be multiple-choice and present 4 options labeled A, B, C, and D.
2. Specify when two correct responses are required and only award full credit if both answers are correct.

PaulDuvall

Category	Term	Definition
Custom Language Models	Amazon SageMaker	A comprehensive service offering tools to build, train, and deploy ML models at scale.
AI Concepts	Artificial Intelligence (AI)	AI is the broad field of creating machines or systems that perform tasks requiring human intelligence.
Supervised Learning	CNNs	Primarily used for image and video recognition tasks in deep learning.
NLP

PaulDuvall

Comprehensive Mapping of AI, ML, Deep Learning, and Related Concepts

1. Artificial Intelligence (AI)

Definition: AI is the broad field of creating machines or systems that perform tasks requiring human intelligence, such as reasoning, learning, and language understanding.

Subfields:

• Machine Learning (ML)
• Natural Language Processing (NLP)
• Generative AI
• Computer Vision

PaulDuvall

prompts for writing article on aws-certified-ai-practitioner-exam

Prompt 1:

Are you familiar with the AWS Certified AI Practitioner exam: https://aws.amazon.com/certification/certified-ai-practitioner/)?

Prompt 2:

I recently passed the AWS Certified AI Practitioner exam and want you to edit this article into a 900-word blog post about my experience. Could you assist me?

Prompt 3:

In the blog, I'd like to reference techniques from the book Made to Stick by Chip Heath and Dan Heath that I used in my preparation, along with how I leveraged generative AI to create practice exams. Here is a link to the prompts I used to generate my practice exams. Overall, the exam was well-structured, but I noticed that scenario questions often had more than one “correct” answer. This made it essential to carefully read the final sentence for context (e.g., “Which is more cost-effective?”, “Which is most secure?”). I also believe that most prep exams don’t fully prepare candidates for these types of q

PaulDuvall

git clone https://github.com/PaulDuvall/cloudproviders.git
cd cloudproviders
sudo chmod +x launch-stack.sh
./launch-stack.sh