Describe AI MLOps/LLMOps Ecosystems and General Use CERTIFIED-AI-EXPERT Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   Which statement best describes the primary goal of MLOps?

   • ATo automate the deployment and monitoring of machine learning models in production
   • BTo replace data scientists with automated tools
   • CTo store all training data in a single centralized repository
   • DTo eliminate the need for model retraining
   Reveal Solution Discussion Select an answer first
2. 7expert · hard

   A government agency is developing an LLM-based system to assist with public records requests. The system must be transparent, and the agency must be able to explain why certain documents are retrieved. The team is considering using a retrieval-augmented generation (RAG) approach. Which practice is most important to ensure transparency and explainability?

   • AUse a fine-tuned model that has been trained on the agency's documents, without exposing the retrieval process.
   • BLog the retrieved documents and the prompts used for each query, and provide a user interface that shows the sources of the generated answer.
   • CImplement a feedback loop where users can rate answers, but do not store the underlying retrieval details.
   • DRely on the model's built-in attention mechanisms to explain its decisions, without additional logging.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A research institute is developing a medical imaging model and needs to track experiments, compare model versions, and share results with collaborators. Which component of an MLOps ecosystem is most essential for this need?

   • AAn experiment tracking system that logs parameters, metrics, and artifacts for each run.
   • BA monitoring dashboard that shows system metrics like CPU usage.
   • CA data versioning tool that only tracks changes to the training data.
   • DA continuous integration server that builds the model code.
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   A news organization wants to use LLMOps to automatically generate article summaries. They need to ensure the summaries are factually accurate and do not misrepresent the original content. Which practice is most directly aligned with this goal?

   • ARelying on the model's training to ensure it never makes factual errors.
   • BImplementing a fact-checking step that compares the summary against the original article and flags discrepancies.
   • CGenerating summaries without any post-processing to preserve the model's original output.
   • DUsing a model with a high number of parameters to improve summarization quality.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A tech company is building a recommendation system that uses both collaborative filtering and content-based filtering. They want to use MLOps to manage the models. Which component is most important for managing the two models together?

   • AA monitoring system that only tracks infrastructure metrics.
   • BA model registry that tracks versions of both models and their performance metrics.
   • CA data pipeline that only ingests user interaction data.
   • DA version control system for the code, but not for the models.
   Reveal Solution Discussion Select an answer first