i dont think this question has a correct answer.
suppose the candidate is off the chain amazing. you know from their git repos, references, etc. that they are qualified and interested in exactly what the position entails. You think they will be a great fit etc.
In my opinion, if you grill that person with coding questions (that you already know they will or will not know based on said page), you may discourage a qualified candidate. With that person I would be honest and open about exactly what the position is and what it entails, the type of team they'll be on, other intangibles that arent as easy to evaluate.
but, if instead the candidate comes for instance with great references and a winning personality, but you have no idea if they are technically competent or not - with that person it would be foolish not to try to determine whether they do or don't have the chops to do the job. you dont want to set them up for failure in a position they cant do; and you dont want to hurt your teams either.
but again, ultimately, I feel that there are many correct answers to this question, the art of it is know which to select in what scenario.
EDIT
oh. i see now. my apologies i phrased it from the perspective of the interviewer. sorry about that. but actually, that answer may still help. Get in your interviewers shoes. Look at your CV, experience, coding ability, math, sales (if there is an client interfacing), etc. If i think i am definitely qualified from a technical standpoint, (like in scenario 1 above), id focus on understanding the firm's clients, their goals, their vision, their teams and people, etc. to see if you are a good fit.
If you have concerns that you are good to go in most areas, but maybe with one exception (like you are solid on all of it except you havent done REST or APIs or what have you extensively before), then I think what you do is, yes, you do prepare by studying specifically the area of weakness then you go in there, and you tell them the truth - which is that you think are are solid on everything except APIs, but you have been reading a book called _____, which actually you found by reading the companies philosophy on ______. So you are confident you can overcome that difficulty.
Then you ask them what concerns they have about your record, if any, and you go from there.
that help?
VAL
Depending on the role. If the job requires 'software development' like skills, you should test coding skills. If the job requires downstream analysis and interpretation of data, you should test more of domain knowledge and analytical skills rather than coding skills.
Thank you for your reply! My question seems not clear, so I explain more above.
The Biostar Handbook can be handy: The Biostar Handbook. A bioinformatics e-book for beginners. . I like what you're suggesting. Perhaps a resource post on this platform where people can share their interview questions and the community can provide their thoughts.
As an interviewer, I rarely have a set of predefined questions. I typically ask interviewees to introduce themselves (45-60 seconds summary of their resumes/CVs) and focus on questions about what they had done in the past, which sometimes can lead to coding questions. What I'm looking for is how they describe and explain bulleted points on their resumes/CVs, and how they justify and defend their work when an alternative approach is proposed.