How do you evaluate skills when hiring AI professionals?

"AI professionals" is too broad of a term. Are we seeking a Machine Learning Engineer responsible for designing and implementing models? Perhaps an ML Ops Engineer to ensure these models seamlessly integrate into our systems for optimum efficiency and scalability? While both of these are AI roles, the actual day-to-day work activities are very different. Regardless of the specific role, however, the crux of the hiring process remains unchanged - a work-simulation-based approach is the most reliable indicator of a candidate's abilities. Mimicking a real-work environment gives candidates the chance to showcase their skills in a practical, relevant setting and gives the most predictive signal of future performance.

To hire top AI talent: 1. Examine research papers, GitHub repos, or online courses for technical proficiency and domain knowledge. 2. Assess collaboration, ethical reasoning, and adaptability through behavioral questions. 3. Use technical and situational interview tests to measure complex and soft skills. 4. Implement the STAR method for structured, effective questioning in interviews. 5. Resume & Cover Letter: Analyze these for qualifications and experience. 6. Simulation & Role-Playing: Use these to assess real-world problem-solving abilities. 7. Validate skills and experience through former employers or colleagues. Prioritize skills and training over merely academic and work experience.

??The rapid pace of technological advancements means that the skills required today might evolve or become obsolete in a few years. I believe we should prioritize candidates with strong foundational skills in mathematics, statistics, and programming. While specific tools and frameworks might change, these foundational skills remain crucial and allow professionals to adapt to evolving technologies. ??However, there might be an initial training period required to familiarize them with specific tools or frameworks the organization uses.

When hiring AI professionals, first ensure they have a strong mathematical foundation in areas like linear algebra and calculus. Check their coding skills, especially in Python and libraries like TensorFlow. It's also crucial they can tackle real-world challenges, so review their projects and portfolios. Ethical understanding of AI and potential biases is a great sign of depth. And remember, teamwork and communication are key; they should be able to collaborate well and explain technical concepts simply. Look for a balance between technical prowess and soft skills. Happy hiring! ??

When defining a job role and required skills, avoid using templates at all costs. Take the time to consider the specific tasks that need to be accomplished and resist the urge to hire a jack-of-all-trades. Instead, focus on attracting candidates that have the exact skills needed for the job. Clearly list the necessary skills for each task to ensure the right fit for the position. List 'What' skills are needed to do 'What' work!

??Online assessments and tests have become a popular method to evaluate the technical skills and knowledge of AI professionals during the hiring process .However, in my opinion, they also come with their own set of challenges: 1. Tests may not capture the full breadth and depth of a candidate's skills and experience. 2. Some candidates might excel in test-taking but lack practical experience or problem-solving skills in real-world scenarios. 3. Over-reliance on tests might lead to missing out on candidates with unique experiences or perspectives that aren't captured in the assessment.

When hiring AI professionals, conducting tests can provide insight into a candidate's technical expertise. These evaluations are essential for sifting through a large pool of applicants and verifying their fundamental domain knowledge necessary for the role. Ideally, these tests should be concise, lasting only 10-20 minutes and focusing on core competencies. However, it's important to note that AI is a vast field, and platforms like HackerRank or Codility may not fully capture the depth and nuances of an expert's experience and skills. For those seeking senior roles, these tests may appear redundant as they have already proven their capabilities through years of practical application.

Overall, Kaggle is a great tool for interviews for AI roles. By using Kaggle, companies can assess the machine learning and data science skills of candidates in a more objective and efficient way. The vast amount of datasets available on Kaggle can be used to test EDA and data processing skills. Additionally, tests that require candidates to train a model on a Kaggle dataset and then predict the outcome of a new data point, can be used to test knowledge of a number of AI algorithms.

Online tools are invaluable for screening AI professionals, but it's crucial to consider both technical and soft skills. Key Points: - Platforms like Kaggle enable you to objectively assess skills such as EDA, supervised/unsupervised techniques, etc. - However, these challenges are time-consuming; engage with candidates beforehand to incentivize participation. - If you have a budget, you can host a hackathon on Kaggle to attract a broader talent pool. - For other roles (e.g. data engineers), MCQ tests on technologies like SQL, Python, Spark, Kafka, etc. are available. While these tools don't offer complete insight into a candidate's capabilities, they effectively narrow down your search, saving time and effort in the hiring process.

Assessments are necessary to evaluate candidates' technical skills and research capability. It also served as a purpose to auto-filter those NOT serious candidates. We are in digital media industry, we always asks candidates first to do a simple research on a metric. Not technical assessment, but assessing candidates' knowledge of the industry. Second assessment (provided after interview) is technical challenging. Candidates are asked to use Colab, to process a given dataset of 2k records (with 140K variables) and list out top correlated variables (against the testing variable). Most candidates withdraw during the second assessment.

While evaluating candidates from their portfolios and projects, I prefer to look into their collaborative works (such as contributions to open-source software) rather than any personal project. AI projects can vary widely in terms of complexity, goals, and methodologies. This lack of standardization makes comparing candidates based solely on their project work challenging. Also, they may not reflect a candidate's up-to-date knowledge and expertise.

Reviewing portfolios is helpful. It shows not just technical skills but also communication abilities. However, the varied nature of AI projects makes direct comparisons difficult. Key Points: - Check the portfolio for domain expertise and its match with your needs. - Look for good documentation and presentation. - Evaluate the complexity and goals of featured projects. This shows problem-solving skills. - Give weight to collaborative or open-source contributions. - See if they've published blogs or opinion pieces on the latest AI technologies. This indicates agility in exploring new developments and assessing their viability for your use case.

??Projects in a portfolio might have been collaborative efforts, making it challenging to discern the specific contributions of the candidate. So, I think we should have a interactive portfolio discussion that allows for a more dynamic and comprehensive understanding of a candidate's work. ??This allows them to elaborate on their roles, the decisions they made, and the impact of their work, providing deeper insights.

AI is a field where individuals must possess the ability to not only solve problems but also identify them. Those who lack an understanding of business and its fundamental processes may struggle to make substantial contributions, even if they are proficient in the technical aspects. Additionally, it’s essential to evaluate someone’s familiarity with the tools that underpin AI. AI doesn’t operate in isolation; it relies on multiple interconnected components to address complex problems.

A candidate's portfolio is a testament to their expertise and journey in the field. While the code provides insights into technical proficiency, the thought process differentiates an outstanding portfolio. An ideal AI portfolio doesn't merely replicate trendy solutions; it echoes the candidate's unique perspective, creativity, and innovation ability. The portfolio extends beyond algorithms or models in today's dynamic AI landscape. It's a narrative of the candidate's problem-solving approach, dedication to open-source contributions, and passion for sharing knowledge with the community. It reflects their hard skills in data analysis and model selection and their soft skills in communication, documentation, and presentation.

In today's world, candidates may give rehearsed answers, which can cloud genuine insights. When interviewing, pay attention to nuances in their responses, as non-verbal cues can provide a deeper understanding of their authenticity. Avoid asking clichéd questions like "What are your future visions for the next five years?" as the job market is constantly changing, and projected plans may be misleading. Instead, focus on their immediate aspirations and motivations, as this can give you more insight into their mindset when actively looking for a job.

??We could make scenario-based questions, pose hypothetical, job-related situations to the candidate, and ask how they would handle them. For instance, "How would you handle a disagreement with a colleague about the approach to a project?" or "Imagine your AI model is not producing the expected results despite multiple tweaks. How would you approach this?" This type of question assesses problem-solving and critical thinking skills.

When hiring AI professionals, assess both technical skills and behavioral competencies. Define key competencies, tailor role-specific questions, and use the STAR method. Evaluate cultural fit, problem-solving, learning mindset, and adaptability. Ask follow-up questions, develop a scoring system, and consider behavioral assessments. Involve diverse interviewers, conduct reference checks, and collect feedback for continuous improvement. This holistic process ensures well-rounded evaluations and informed hiring decisions. Also consider using behavioral assessment tools or personality assessments to complement your interview process. These can provide additional insights into a candidate's behavioral traits.

I totally agree with this, behavioural and situational questions helps in better understanding of candidate’s exposure to various challenges, teamwork, problem solving and knowledge towards specific areas especially when hiring for specific AI roles. Compliance and risk assessment are also a very critical part of AI related role and processes.

Engaging in a multi-tiered review process involving different team members across two to three sessions can provide varied perspectives on the candidate's suitability for the role. Here's a suggested approach: Tech Talk: Dive into a discussion on technical problem-solving to measure domain expertise. Peer Vibes: A relaxed interaction with potential teammates evaluates cultural fit and collaborative spirit. Mindset Mapping: Understand the candidate's core values, ambitions, and expectations. Employers can make more informed and holistic hiring decisions by staggering evaluations and pooling diverse viewpoints.

I agree with these thoughts and feel the new wave of cliched questions is asking things based off rote memorization. After truly partnering with AI, all of the syntax based questions have become irrelevant because using AI requires a whole new way of approaching problems. It alleviates the need for mundane technical details and opens the door to problem solving end to end. It's like when people ask for directions -- there are those people who are used to memorizing things but then there are those of us who know the google algorithms are going to know more so dont memorize the routes, we learn to leverage the data...

??We could also organize sessions where candidates collaborate with potential future colleagues on a task or problem. This provides insights into their teamwork, communication skills, and adaptability to feedback.

Recruiting the best talent, especially in AI, goes beyond mere technical prowess. It delves into soft skills, which are increasingly becoming non-negotiable. A technically proficient AI expert lacking team synergy can do more harm than good. This becomes even more pronounced in remote roles where organizational abilities and individual discipline. Get personal, but remain professional. Inquire about their reasons for job transitions or perhaps a challenging experience from a past role. Such questions uncover personality traits and gauge cultural fit and resilience. Remember, every hire is an investment in the company's future. Hiring isn't merely a task – it's a journey to find the right fit.

While AI is indispensable and readily available to everyone a part of as short length courses and advent of ChatGPT. The competition becomes tough to hire the right AI talent. What separates right AI talent from the crowd is their ability of problem solving and understanding implications of AI project they are building. It is not a usual day when one has put their SWE skills to practice along with having mathematical and research paper knowledge of those AI models. The online engagement through blogging, research paper publication, YT channel, GitHub profile and portfolio websites can help gauge their involvement in topics related to AI.

To evaluate AI professionals for hiring, consider technical skills (e.g., Python, ML algorithms), problem-solving abilities, communication (technical and non-technical), and collaboration skills. Assess candidates through interviews, coding challenges, and real-world problem-solving examples. Additionally, seek candidates passionate about AI and eager to learn. Involve various stakeholders in the hiring process to ensure a comprehensive assessment and maintain objectivity, avoiding biases.

Focusing on a candidate's project portfolio, contributions to open source, and exposure to cloud-based model deployment are essential aspects when evaluating AI professionals. These factors demonstrate practical experience, hands-on skills, and a commitment to staying current in the field. Additionally, a strong understanding of model statistics is vital for effective data analysis and model performance evaluation, making it an important consideration in the hiring process.