Enterprises are incessantly improving their recruitment endeavours to seek out and onboard data scientists who can assist in analysing patterns, leveraging analytics, and identifying factors and insights that influence product strategy. Data scientists play a key role in the broad implementation of machine learning and AI methods to elevate metrics such as engagement, retention, and profits.
Despite the escalating need for data scientists, many enterprises do not possess a definite hiring plan to find and recruit skilled data scientists.
In this article, Jeremy Stanley, the Chief Data Scientist and EVP of Engineering at Sailthru, dissects his organisation’s hiring methodology for data scientists.
Outlined below are the crucial points to remember:
The basis of your recruitment process
If you have a recruitment system in operation, here are three factors to bear in mind to enhance your existing hiring approach.
- Augment the precision and success ratio of identifying exceptional candidates and persuading them to accept your job offer.
- Minimise the probability of good candidates withdrawing from your recruitment process prematurely.
- Guarantee that the recruitment process does not consume an excessive amount of your time and resources.
In addition to these factors, it is vital to establish a robust base and construct a set of fundamental principles to guarantee that you attain all of your objectives.
Listed below are the essential recruitment principles that you should adhere to:
Ensure that your recruitment approach is continuous.
Establishing a recruitment process that can be maintained over a prolonged period is crucial for any flourishing organisation. The objective should be to make your recruitment practice a continuous priority rather than a temporary one. Dedicate sufficient time and resources to persistently search for, evaluate, and recruit the finest and most competent candidates, regardless of the season. By generating an ever-evolving and steadily enhancing recruitment campaign, you can ensure that your business will have the appropriate personnel in the appropriate roles.
Instead of using generic interview queries, it is crucial to customise your questions according to the specific role you intend to occupy. This will enable you to gain a more in-depth comprehension of the candidate’s qualifications and capabilities and offer you a clearer perspective on whether they are the suitable choice for the position.
Ask questions that are particular to the duties you anticipate the candidate to carry out once they become part of the team.
“Comprehend the problems your team is presently confronting and the obstacles you anticipate triumphant candidates to encounter,” suggests Stanley.
To diminish your prejudices, establish an unprejudiced recruitment process.
Robust programming and numerical abilities are necessary for recruiting a data scientist. Therefore, initiate your interview by examining these competencies.
After evaluating candidates on their technical abilities and qualifications, you can proceed to assess their interpersonal skills, such as their proficiency in communicating effectively and their problem-solving proficiencies. Eventually, you can evaluate their compatibility with your organisation and how effectively they would work as a component of your team.
It is crucial not to jump to a conclusion about a prospective candidate merely based on their compatibility with your firm. Doing so may give rise to a biased viewpoint that could result in the premature rejection of a feasible alternative.
Establish a organised recruitment procedure to make an excellent first impression.
Most recruitment procedures involve several interviews with diverse individuals, where identical queries are posed. After the interviews, the candidates must wait for a specific period before receiving feedback on their application, and frequently, their inquiries might go unanswered.
Your recruitment method will be their initial encounter with how your company functions, so you should have a well-defined plan in operation.
Offer potential employees with an all-encompassing outlook into your corporate culture and the challenges they might confront while collaborating with you and the rest of the team. This will provide them with a better notion of what to expect after joining your company.
Participate with the team in decision-making.
Maintain a well-defined and palpable structure for assessing candidates.
Furthermore, if you envisage that the data scientist will have to collaborate with other departments, it is vital to incorporate pertinent key stakeholders from those teams during the ultimate election procedure.
Speed up your recruitment process.
Although it is advantageous to take sufficient time assessing your candidates, you run the risk of losing out on good potentials to your competitors.
To guarantee an uninterrupted and productive recruitment process, it is imperative to accelerate the transition of potential candidates across the distinctive phases. Investing in appropriate tools and infrastructure that can monitor and analyse the numerous stages of the recruitment process can aid in identifying areas that necessitate further enhancements to sustain the impetus.
Interview Phases
An interview for a data scientist entails six stages.
Pre-screening Stage
To review each candidate’s CV, prior experiences, and qualifications, one can access them here. However, to evade interviewer bias, one may choose to skip this stage and instead dispatch a take-home test to all potential candidates.Take-home Test
This short test evaluates the candidate’s proficiency in resolving issues in their selected domain.Conviction Stage
This is the phase where you convince candidates who have cleared your take-home test to attend the interview.Data Gathering Day
To proceed with the selection process, we will request potential candidates to participate in a full day interview. Throughout the interview, we will present them with an open-ended challenge and ask them to generate an appropriate solution and present it to us. This section of the interview will permit us to evaluate the candidate’s abilities in a simulated and regulated setting.
Preparation is the crucial element in facilitating this process- Prepare a set of printed instructions beforehand to hand out to them as soon as they arrive.
- Outfit each candidate with a laptop that has pre-installed tools and software programs required to perform their tasks. This should incorporate HomeBrew, Eclipse, Anaconda (Python distribution), RStudio, R, Emacs, Vim and Java 7. This will guarantee that the candidate will readily have all of the essential data and programs at their disposal.
- Solicit feedback from candidates and integrate it into your process to fulfil the requirements of your company.
Final Decision
Stanley recommends utilising these five factors as the foundation of your hiring determinations.- Problem Structure What suppositions did they assume, and what tactics did they use to confine the problem’s scope?
- Technical Proficiency. How legible, reliable, and flexible is the submitted code?
- Analytical Proficiency. How rational is their rationale for the approach they proposed?
- Communication Proficiency. How proficiently did they articulate their work, tactics, approach, and conclusions?
- Value of their Work. What would be the worth of the work of your new hire after it is refined and implemented?
Communication
The final phase of the process is to communicate directly with candidates. It is crucial to provide feedback to Data-day candidates, even if the result is negative, so that they can take away something worthwhile from their experience. Following up with candidates who were not sent an offer letter is crucial to this process.
Implementing a well-structured hiring process can be advantageous in identifying potential candidates whose qualifications may not be readily evident from their resumes or applications. This comprehensive approach removes any guesswork from the hiring process and instead equips you with the necessary data to make an informed decision.
