Tool Selection Based on AI Project Requirements AIE Practice Questions (Page 6)

Questions 26–30

1. 26expert · medium

   A telecommunications company is building an AI model to predict network failures. The data is generated continuously from network sensors and is extremely time-sensitive. The company has an existing on-premises data infrastructure and a team of data engineers who are proficient in SQL and Python. They need to build and deploy the model quickly, but they also have a mandate to reduce infrastructure costs. Which approach best balances speed, cost, and the time-sensitive nature of the data?

   • AUse the on-premises infrastructure for both data processing and model training, but only run training during off-peak hours to save costs.
   • BUse the existing on-premises data infrastructure for data processing, and use a cloud-based ML platform for model training and deployment, with a streaming pipeline to send relevant features to the cloud.
   • CUse a cloud-based streaming analytics service to process all sensor data and train the model entirely in the cloud, without using the existing on-premises infrastructure.
   • DBuild a completely new on-premises ML platform, including new hardware, to keep all data processing and model training in one place.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   A startup needs to quickly provision AI infrastructure without upfront hardware investment and wants to scale resources up and down based on demand. Which deployment option best fits these needs?

   • AOn-premises tools
   • BLocal desktop software
   • CCloud-based tools
   • DHybrid deployment with on-premises only
   Reveal Solution Discussion Select an answer first
4. 28expert · hard

   A company is building an AI system that requires continuous learning from streaming data. The system must process data in real time and update the model without downtime. Which tool selection best supports this requirement?

   • AUse a cloud-based AutoML service that retrains on a schedule.
   • BUse a batch processing tool and retrain the model nightly.
   • CUse a single monolithic application that handles all tasks.
   • DUse a streaming data platform, an online learning framework, and a model serving platform with rolling updates.
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A mid-sized company is evaluating two ML platforms: an open-source tool with a steep learning curve and a commercial platform with a higher cost but better support. The company has a small data science team that is already familiar with the open-source tool. Which factor should be the primary consideration in the decision?

   • AThe popularity of the tool on social media.
   • BThe number of features in each platform.
   • CThe color scheme of the user interface.
   • DThe total cost of ownership, including training time and support.
   Reveal Solution Discussion Select an answer first
6. 30expert · medium

   A software company is choosing between two AI tools for a natural language processing project. Tool A is open-source with a large community and frequent updates, but requires significant in-house expertise to configure and maintain. Tool B is commercial with strong vendor support and a user-friendly interface, but has a high annual license fee and limited customization options. The company has a small team with limited NLP experience but a healthy budget. Which tool should they choose?

   • ATool A, because open-source tools are generally more flexible and have a larger community for troubleshooting.
   • BTool B, because the team's limited NLP experience makes vendor support and ease of use more critical than cost or customization.
   • CTool B, because commercial tools are always more reliable than open-source tools.
   • DTool A, because the company has a healthy budget and can hire external consultants to handle the configuration.
   Reveal Solution Discussion Select an answer first