It was a HireVue-based online assessment. There were a total of 8 questions: 3 coding, 4 video responses, and 1 location preference.
The assessment started with the location preference, followed by 3 video responses. Then, you would encounter 3 coding questions with different time allocations. The last question was another video response to explain the logic of one of the coding questions.
Location Preference
What is the position? Why are you a good fit?
Which new tech did you teach yourself? What were its pros and cons?
How will you find product names from a large set of Twitter feeds? Which ML algorithm will you use? Explain the process.
Given a tokenized sentence, extract the words and create a sentence.
Explain how you approached coding the tokenization question.
Sentence in Reverse Order:
QUESTION TOKENIZATION THE CODING APPROACH YOU HOW EXPLAIN
ALGORITHM ML WHICH USE WILL YOU FEEDS TWITTER OF SET LARGE A FROM NAMES PRODUCT FIND WILL YOU HOW
CONS AND PROS ITS WERE WHAT YOURSELF TEACH DID YOU TECH NEW WHICH
FIT GOOD A YOU ARE WHY
POSITION THE IS WHAT
The following metrics were computed from 6 interview experiences for the IBM Cognitive Software Engineer role in United States.
IBM's interview process for their Cognitive Software Engineer roles in the United States is fairly selective, failing a large portion of engineers who go through it.
Candidates reported having very good feelings for IBM's Cognitive Software Engineer interview process in United States.