- (Topic 2)
You have deployed a new public web application that allows users to register and login with email ids, phone numbers, or user ids. You are seeing some unusual activity with user registrations and logins from a few IPs. A large number of accounts were created very quickly. Logins are also hap-pening quickly thereafter from these new accounts. Different parts of the application are being ex-plored, all of which are putting a heavy load on the application. What could be a problem and how can you solve it?

1. A. A hacker group has hired a bunch of people to create accounts and manually use the syste
2. B. Use Cloud Asset Inventory to see if there have been changes in the inventor
3. C. Bots are creating accounts and then using the
4. D. Use Google Cloud's Web App and API Protection (WAAP).
5. E. Bots are creating accounts and then using the
6. F. Use Identity-Aware Proxy to re-strict the users to known users.
7. G. Automated testing tools might still be running and creating account
8. H. Use Identity-Aware Proxy to restrict the users to known users.

Correct Answer: B
Bots attacking the application is the most likely scenario in this case. Using WAAP is the right protection plan: Anti-DDoS, anti-bot, WAF, and API protection help you protect against new and existing threats while helping you keep your apps and APIs compliant and continuously available.
https://cloud.google.com/solutions/web-app-and-api-protection

- (Topic 1)
Your company provides car maintenance services. It is conducting an internal hackathon to identify new ideas that could expand their business. The teams have pitched different ideas and have started working on it. They have to present their application to the judges within 48 hours. A presentation alone is not enough; they have to demonstrate a working proof of concept. The team that you are mentoring is going to recommend additional ser-vices to drive in customers based on the brand of car they drive in. They need to be able to identify what brand of car the customer has, based on a photograph automatically taken at entry. They have already discovered an open source database of car images collected by online enthusiasts. How should they implement this solution?

1. A. Use Deep Learning Containers that are preconfigured and optimized containers for deep learning environments.
2. B. Use AutoML Image - upload the images and let it create a working model for you.
3. C. Use TensorFlow to create a model that will identify the car brands; use the available data to train the model.
4. D. Use Cloud Vision AI that is able to detect logo
5. E. Write only the code to integrate in-to your workflow.

Correct Answer: B

It would be most straightforward to use AutoML Image. Put the images in Cloud Storage, point to it from AutoML, and start the model building process.
Reference Link- https://cloud.google.com/automl

- (Topic 2)
DriveSuper Inc. teaches its clients to drive cars and bikes and helps them get their license. They are planning to build a mobile application where users can sign up, plan their schedules, and take stock of progress. They want the onboarding process to be smooth and frictionless, giving users a great experience from the get-go. They want this done as quickly as possible and not be expensive. What is their best option on Google Cloud?

1. A. Build the mobile app with Cloud SQL as the backend
2. B. Build the mobile app with Cloud Storage as the backend
3. C. Build the mobile application with Firebase as the backend
4. D. Build the mobile app with Cloud Spanner as the backend

Correct Answer: C
Firebase/Firestore is easy to build and is suitable for user information that could bvary in nature.

- (Topic 2)
Your client is a financial services company giving loans based on customer profiles. As part of the regulatory compliance, they have to collect a bunch of different documents with know your cus-tomer (KYC) information. They want to be able to process the information in these documents quickly and at scale. They want to integrate the chosen solution as quickly as possible. What are your options on Google Cloud?

1. A. Integrate the Cloud Vision API to create a custom model to handle the documents.
2. B. Create a model using TensorFlow and integrated it into the process workflow.
3. C. Integrate the Lending DocAI and Document AI in two there processes workflow of the processing loan requests.
4. D. Integrate the Natural Language API to read the request sent in by clients and to process the forms.

Correct Answer: C
Lending DocAI is a pre-packaged AI solution that speeds "up the mortgage workflow processes to easily process loans and automate document data capture, while ensuring the accuracy and breadth of different documents (e.g., tax statements and asset documents)."
https://cloud.google.com/solutions/lending-doc-ai

- (Topic 1)
Your organization needs to plan its cloud infrastructure expenditures. Which should your organization do?

1. A. Review cloud resource costs frequently, because costs change often based on use
2. B. Review cloud resource costs annually as part of planning your organization’s overall budget
3. C. If your organization uses only cloud resources, infrastructure costs are no longer part of your overall budget
4. D. Involve fewer people in cloud resource planning than your organization did for on- premises resource planning

Correct Answer: A
Review cloud resource costs frequently, because costs change often based on use because One need to know current usage/ trend for planning; While public cloud eliminates capex, and gets into pay as you go model, the usage pattern determines the cloud cost and hence needs to be measured frequently to enable better forecast