The result values of all of questions are graded by the system. I'm not sure why you would want a human to do it.
Let me remind you about the exam format from http://cloudera.com/training/certification/cca-spark.html
Exam Question Format
Each CCA question requires you to solve a particular scenario. In some cases, a tool such as Impala or Hive may be used. In other cases, coding is required. In order to speed up development time of Spark questions, a template is often provided that contains a skeleton of the solution, asking the candidate to fill in the missing lines with functional code. This template is written in either Scala or Python.
You are not required to use the template and may solve the scenario using a language you prefer. Be aware, however, that coding every problem from scratch may take more time than is allocated for the exam.
We are not requiring you to use a specific tool, nor to use a specific programming language. From this, you may surmise that we are looking at the results you generated for your solution, not the code or process that you used to get to that solution. I don't think you want a human looking through your sqoop results to make sure you got the correct 25 million rows, it might take a while.
I wish the grading was done by humans, I would have passed the exam since we would have been graded by the fact whether we are providing the solutions correctly, even if it took a while. Since it's a new pattern of exam and not all of us have gone through this procedure before and it's expensive (as I'm from India) , this process would have been benefiting.