HOTSPOT - A company wants to build generative AI applications by using Amazon Bedrock. The company wants to minimize development effort. Select and order the model development techniques from the following list from the LEAST development effort to the MOST development effort. Each model development technique should be selected one time.
Answer is in the explanation below.
Reference / correct answer:
Least Effort: 1. Prompt Engineering - Requires no pre-trained model or any data preparation. 2. RAG - Only sets up the data sources, no pre-training is required on the datasets. 3. Fine-tuning - Requires labeled datasets and a training job. Most Effort: - Continued pre-training - Resource and effort intensive process as the datasets need to be trained on a large and domain-specific tasks. upvoted 1 times
Explanation Prompt – Least effort Uses existing foundation models as-is. No data preparation or training required—just prompt engineering. RAG (Retrieval-Augmented Generation) Requires setting up a data source (e.g., vector database, embeddings) but does not retrain the model. Fine-tuning Requires labeled datasets and a training job to adapt the model to specific tasks or styles. Continued pre-training – Most effort Involves training the model further on large, domain-specific corpora. This is the most resource- and effort-intensive approach. upvoted 3 times
1. Prompt 2. RAG 3. Fine-tuning 4. Continued pre-training upvoted 2 times
Question 38(Topic 1)
A company is developing a product recommendation application by using a generative AI model. The company must minimize the application’s environmental impact. Which solution will meet these requirements?
Select an option, then click Submit answer.
○
Optimize the deployed model architecture to prioritize computational efficiency during model inference.
○
Adopt a distributed inference approach by using multiple smaller models across multiple Availability Zones.
○
Adopt a hybrid strategy by deploying the model on premises and storing the data on AWS.
○
Deploy multiple models and use a dynamic model selection mechanism that queries different models randomly.
Reference / correct answer:
Optimize the deployed model architecture to prioritize computational efficiency during model inference.
Most accepted answer: A. Optimize the deployed model architecture to prioritize computational efficiency during model inference.
Community votes: A=2
Selected Answer: A The goal is to minimize environmental impact. The biggest contributor to environmental impact in AI systems is often: 1. Compute usage 2. Energy consumption 3. Resource utilization during inference and training upvoted 1 times
Selected Answer: A The correct answer is A. Optimize the deployed model architecture to prioritize computational efficiency during model inference. In the context of the AWS Well-Architected Sustainability Pillar, the most effective way to reduce the environmental impact of a generative AI application is to reduce the total energy consumed during the inference phase, which often accounts for the majority of a model's lifetime carbon footprint. upvoted 1 times
Question 39(Topic 1)
HOTSPOT - A company wants to build a new ML solution. The company already has data. The company needs to understand the ML lifecycle before building the solution. Select and order the steps from the following list to correctly describe the ML lifecycle. Select each step one time.
Answer is in the explanation below.
Reference / correct answer:
If the data is already contained then: 1. Prepare the data (Data Wrangler tool for instance). 2. Train the model (Fine-tuning etc) 3. Test the model 4. Deploy the model (SageMaker pipelines for instance) upvoted 1 times
nothing difficult here. just think the logical steps needed. 1. Prep the data 2. Train it 3. Test it 4. Deploy it upvoted 1 times
Limited Time Offer – Save Now!
Don’t miss out — get full access at the best price.
