By AI Trends Staff
Skilled AI workers are being urgently sought. Knowledge and some experience in artificial intelligence and machine learning are the job skills most in demand in 2021.
That is confirmed in a survey by Hackerearth of 2,500 developer recruiters and hiring managers reported in a recent account in The Enterprisers Project.
The AI field is expansive, with a wide range of skills represented. “This is an incredibly broad field, and not all jobs will require the same skills. As your organization competes for talent, don’t let enthusiasm cloud your judgment,” stated Rajan Sethuraman, CEO of LatentView Analytics, author of the piece. The company offers analytic services.
The interviewer is looking for an understanding of whether a candidate can translate their AI skills into business results, he suggested. He offered some interview questions he has tried recently to prospective AI hires:
What is your experience with AIOps or MLOps?
AIOps is short for AI for IT operations. MLOps is a practice for collaboration and communication between data scientists and operations professionals to help management production in the machine learning lifecycle.
Work with organizations that have invested millions of dollars into AI initiatives have taught Sethuraman that boards of directors are asking how the investments in AI have led to real growth. “The focus is now on holding AI accountable and how to truly operationalize AI,” he suggested. “It’s about better operations, not big new projects.” As a result, AIOps has jumped to the front of desirable skills for AI.
AIOps is standardizing AI pipelines, generating in-depth performance analytics, and automating workflows. This helps teams to improve algorithm performance using real-time metrics. Sethuraman also looks for some knowledge of application performance monitoring (APM), which can examine a range of metrics. “Their performance-driven background orients the initiative toward measurable business objectives,” he stated.
What are some novel use cases for AI in X field?
If you are applying for a job in banking, you might be prepared with some relevant examples. Sethuraman might be impressed by a technical expertise with relevant algorithms. “In the case of loan approval, for instance, the KNN algorithm is a good Supervised Learning algorithm because it effectively sorts applications into two classes: approved and disapproved,” he stated.
He will be looking for the answer to translate from technical to practical. He might ask how the candidate knows the training data is not biased, and how advances in explainable AI might help justify automated loan decision-making if the results are challenged. “ If they are thinking through these implications, you know that they understand your business and the potential challenges they will encounter on the job,” he suggests.
What emerging areas of AI are you excited about and why do they matter for our business?
Here the interviewer is probing to see if the candidate has done some thinking and imagines himself or herself in a role, what that is, and how it fits into the company’s technology trajectory. “It also shows that they are keeping up with their field and will likely continue to do so at your company,” Sethuraman states. Employers are looking for new hires to be flexible and have a desire to upskill, important in a cutting-edge field like AI.
He cautions to be on the lookout for candidates who want to extend certain technologies or areas of AI research because they have experience or a personal preference. “You need to vet this in the interview,” he suggests, to ensure the candidate is pragmatic and focused on relevant projects.
Interview Questions for AI Job Candidates Span Likely Scenarios
Another helpful guide to AI interview questions has been published on the blog of Edureka, a company offering online training from its base in Bengaluru, India. The list was collected after consultation with AI certification training experts and is divided into three sections: beginner, intermediate and scenario-based.
At the basic level, one question is: what’s the difference between AI, machine learning and deep learning? The answer is provided in a three-column table format.
Another question asks the interviewee to provide an example where AI is used on a daily basis. The answer provided talks about Google’s search engine, and how it uses predictive analytics, natural language processing and machine learning to recommend relevant responses. It bases its recommendation on data Google collects such as the respondent’s search history, location, and age.
An intermediate question is, what is better for image classification, supervised or unsupervised classification. The answer is in supervised classification, the images are manually fed and interpreted by the machine learning expert to create features classes. In unsupervised classification, the machine learning software creates feature classes based on image pixel values. Therefore, it is better to choose supervised classification because it will be more accurate.
A scenario-based question asks the candidate to show the working of the Minimax algorithm for choosing the next move in a Tic-Tac-Toe game. Two players are involved, one trying to get the highest possible score and the other trying to get the lowest possible score.
The answer provided shows diagrams, formula, and text and ends up with a formula justifying the recommended first move.
This PhD in Quantum Computing Put Some Thought into His Future
A recent example of a young computer scientist thinking ahead was the story published in The Guardian of Zak Romaszko, a 27-year-old student who recently earned a PhD in quantum computing from the University of Sussex. His undergraduate degree is in physics from the University of Liverpool.
The quantum computer, which uses quantum bits rather than regular bits used by standard computers, are seen as the key to solving very complex problems in a manageable amount of time. While still very much in the research stage, its development is being pursued by many top science teams and big tech companies including Google and IBM.
“It will be able to solve problems that might take computers millions and billions of years in timescales that are more realistic to humans,” stated Romaszko. “It seemed to be that this would be the way forward in how big calculations would be done in the future.”
He spent four years on the project as part of the university’s Ion Quantum Technology group, graduating in June 2020. He now works as a microfabrication engineer for a spinoff company founded by his university advisor Prof. Winfried Hensinger, called Universal Quantum, which is working on commercializing the technology to make a large-scale quantum computer. The company raised $4.5 million last June.
Romaszko is happy to be pushing the boundaries of computing. “With a PhD you’re basically learning about a field and a very narrow area of science that you just plan to push out a little bit further and expand human knowledge. It’s really exciting,” he stated.
Perhaps the message for those interested in working in the field of AI, is to imagine yourself in that future role. It can happen.
Read the source articles and information in The Enterprisers Project, on the blog of Edureka and in The Guardian.
This post was first published on: AI Trends