Episode 112

Rommy Ghaly on Evaluating Technical Skills vs. Cultural Fit in Hiring, Striking a Balance for Effective Team Dynamics

TIM: We're live on the Objective Hiring Show with Rommy. Rommy, welcome to the show. Thank you so much for joining us.

ROMMY: Thanks very much, Tim. Thanks for having me. Yeah,

TIM: And on a public holiday, no less. So I think that's a bonus points for you. We're off to a good start.

ROMMY: yeah, it's my pleasure.

TIM: And where I'd love to start today is just to hear a little bit more about yourself, just so our audience can start to build up a picture in their head of who they're listening to.

ROMMY: Great. Yeah, absolutely. My name is Rommy. I'm the head of data at CircleCI. CircleCI is a continuous integration, continuous deployment delivery platform that developers use to build, test, and deploy code in a way that is very continuous, obviously from the name. I've been working in data for over 20 years. I I cut my teeth. In data specifically at eBay back in 2004, fall of 2000 and spring of 2004. So it's been over 20 years. And I've been in data ever since. For over 20 years, I have been working specifically in data because I love it, and it's something that I find is always interesting and something that I think is absolutely a long-term career path. I think when I started out, everyone saw data analysts as being an entry-level position, but as we know now, much later data is a full-fledged long-term career path in the truest sense. So it's something that I've been very proud to have done and really enjoy.

TIM: Yeah, you couldn't have timed it better with what's happened in the last 20 years. Like, I can imagine things have changed a lot from when you started to now. What do you love most about working in data?

ROMMY: Every day is a different problem. You're measuring a different problem. You're solving a different problem. There are many kinds of data, people, and you're going to say data, and I'll say data. It's going to be a thing. I'm sure you've experienced this before. But for me, there are different kinds of data people. There are those who are very scientific, right? Like Yann LeCun from Facebook or Google, like the big data scientists who are building foundational LLMs, right? But then there are the operational data people. And I put myself more in that bucket, right? I'm familiar with how the more scientific side of things works. But really my thrill is being able to come in every day, trying to answer a question with data, then figuring out how to answer that question, then delivering the results and the story back to the leadership, the stakeholders, and the money people to make the decisions. And to me, every day is a new thing, every problem is a new problem. Is what makes data really such an exciting and fun place to be, and the cycle times for analysis are quite short too; you're not building a massive feature or a massive part of your product. To me, the joy of being able to solve and get a quick answer to a quick question. Or to maybe do a little bit of a deeper analysis can be really interesting, but these things don't usually take months. They take days to weeks, right? Even shorter if it's a smaller question to answer. So to me, it's always exciting, and it's always interesting, and there's, you're never bored in data.

TIM: It sounds like running a company actually every single day is a little bit different, a little bit new, exciting, and sometimes quite stressful. Sometimes a lot is being thrown at you at the same time, but I guess we wouldn't change it for the world.

ROMMY: That's right. That's right. We do it because we love it.

TIM: Exactly. Now you mentioned you've worked in data analytics, called what you like, for 20 years. Now you've aged remarkably well. I have to say that for our viewers who are our listeners who are listening, not watching. What's the biggest change you've seen in the way candidates are evaluated for these roles? Has it changed at all in those 20 years?

ROMMY: It's a super interesting question. I think it has, when I was, when I started off doing it, no one knew how to analyze data. It was crazy, right? I got to travel and live all over the world because of the jobs that I got, right? People were, there was never enough data, people. And there were always too many jobs and too many people, too many companies that needed data folks. So that's probably the biggest thing that has changed. Now it's hiring; it feels like 50 percent of people coming out of university right now are data people. That's how crazy it feels. So that's, I think, really the biggest change is where it started off when, when I started off back in 2004, our tech stack was very different. We were using Teradata at eBay. We were one of the biggest customers at the time. We were using MicroStrategy and Business Objects, if anyone remembers those tools, and enterprise data warehousing was still super new. And now there are a dozen different tools for every part of the data stack. So I think for me it's really been, it's really been just how the industry has evolved and how there are so many more people doing data. And so hiring for data has become a very different thing. And now there's like different flavors of data, right? There's data science, ML, and AI; there's analytics and engineering; there's data engineering; there are people building; there are like the deep learning scientists, the LLMs, and the generative AI side of things. There are so many different parts of data now that exist that didn't exist before. So I think it's just, it's evolved incredibly, and the people that work in it have evolved incredibly.

TIM: I'm struck by a quote from one of our. previous guests that I just put on LinkedIn yesterday. He said something to the effect of, 10 years ago, if you built a dashboard, you were the company's hero. Now you're just another analyst that doesn't understand the business context. Just harsh, but funny and probably accurate. Has our expectations over people's skills change is the average data person now just fundamentally more skilled because the industry is a lot more mature.

ROMMY: Yeah, table stakes are way different now than they used to be like when I started like sequel, knowing sequel was you were you were a genius, right? And so just the ability to type sequel, get a result set and chart it was seen as. As magic back then, right? And we've seen the evolution of Python since then, right? It was Python was not an analytical tool back in 2004. You've seen that change. And now it's self serving is a thing where people can. Everyone can do a pivot table now back then it was like finance people and data analysts could do pivot tables. That was pretty much it. And now everybody can do pivot table. Everybody who has access to data can build their own dashboards tools like Looker and Tableau and and a lot of these other tools are enabling self serve in a way that has never existed before. So pretty much anyone can create a dashboard, but the fundamental thing that, that hasn't changed is. Data literacy is still not widely understood. People are more data literate than they used to be, but you can't just give raw data to somebody and tell them to build a dashboard and expect that they'll build it correctly, because understanding the underlying data and how the data is working is a very critical thing. So the way you aggregate data is super important. I think the broad question is, will AI take over the data analyst role? Everybody's questioning their own jobs nowadays; people are doing it about data. I don't think so. I think a lot of data is in how you take data and understand it. Tell the story with it and make decisions with it. And I think that an AI giving you a result for a question you asked for is going to give you about 3 percent of what you actually need to make that decision. Then you need to go do the deep digging and understand why that data is. I think, to go back to your original question. The data analyst role goes far beyond what it used to anymore. And I think that's a great thing. We become strategists, we become consultants, and we become thought partners to folks in companies. And that's wonderful. That's a wonderful thing.

TIM: Where do you see the blend of skills that the typical analyst will need in the next couple of years? Is it the case, for example, that with the development of your CLAWs, your Chats, and your PTs, maybe that coding skill set is becoming arguably less relevant? Proportionally, and maybe in a year's time, it is more of a, either almost like a domain expert who wasn't even a data person but knew a lot about the CX or the product or whatever, who now has this tool, which allows them to be an analyst in a sense, or more of these kinds of really strong communicators that are almost salespeople. Like, where do you think it's going to go?

ROMMY: It's a really good question. I think for me personally, I always looked for the communicators, the storytellers, the people who can connect the dots between data and business results and decision-making, so for me, it really hasn't changed that much. And I think that what AI helps us to do, and I think this is going to be the case, whether you're a data person. Or an engineer, or even to a certain extent, a designer, it's going to help reduce the bloat and the churn of fine-tuning a query. Or, the kinds of things that, like parsing a reg X, right? This was the first thing I did with ChatGPT back in the day. It was like, I had this huge string, and I was trying to extract one thing from a JSON blob, and it was just super annoying, and I popped it into ChatGPT, and it spit out the exact SQL I needed to do that. That's the kind of thing that I think it's going to help out with, like taking natural language questions and. No AI is going to tell me the business decision I need to make. They'll tell me the answer to my question, but they will not be, an AI won't tell me what I ought to be doing as maybe not yet. I don't want to say never. But to me, what I look for when I hire data folks. Is that thing that AI can't do well, which is how do I not just answer the question, but how do I take that answer to the question and suggest a path forward for leadership for the business. And I think that was always a critical skill for data analysts. And I think with AI, we just like. It helps us get rid of the drudgery of the things that take up time that can be automated, and it allows us to focus more on, on telling the story and driving the decisions.

TIM: I'm trying to think back to, so the last role I had. In analytics myself was about six years ago was in hotel price comparison website And i'm thinking back to our team of analysts and our maybe relative strengths and weaknesses and I wonder if then The expected domain knowledge is going to have to be so much higher because surely that domain knowledge and business acumen Factors that are going to drive how impactful your analysis or thought is. And I'm thinking back to us, six analysts who probably had never stepped foot in a hotel from an analytics perspective ever. We were operating in like an online environment and thinking about things almost like a layer away from the actual core customer of in a hotel, making a booking, going on a holiday. Like we were so abstract from that, if I don't wonder whether that's going to have to change a lot.

ROMMY: Yeah, absolutely. One of the things I You know, in hiring and seeking out analysts that I've, that I'm doing more so today than I've done previously is folks who are really, you have to be super curious, right? It's not just about, taking a data ticket, coming up with analysis, delivering the, analysis and closing the ticket. I want people to think about the prop, the question that the person is trying to ask, or the question that the business is trying to ask and to come up with the. Based on their understanding, their thorough understanding of how you make money or how the product works, or, how we report to the board or to the street. It's not about delivering an analysis. It's about understanding the analysis in the context of the business and what the business is trying to do. So one of the things I always hire for is this person is extremely curious, and will they have an opinion? And there's a belief. Maybe there is a belief. I don't know. I always sense that there's a belief out there that data folks are not people of opinions. There are people of facts, right? That there are people who deliver numbers. But I think having an opinion is probably the most important thing for a data analyst because you can. Answer: You can use data to answer questions any number of ways, but what is the way that is the best way? The way that's going to drive the impact that you needed to drive. And not many people are great at that. And that's really what I look for is people who have a menu of different things that they can do, but they pick the thing that they believe is going to drive the most impact and answer the question in the best way possible.

TIM: So you're looking for people with opinions, but hopefully ones I assume are still grounded in reality. And if they find an answer that is contrary to their opinion, they're going to hopefully change it.

ROMMY: Right. You go into everything with a hypothesis, and it's an informed hypothesis, right? And the goal of data is to prove or disprove your hypothesis. And so to me, that's exactly right. It's, you go in with an informed opinion, and you come out with some basis in fact, right? And that's the goal I think of for all analysis.

TIM: Have you ever worked in an environment where I was a little bit backwards, where it's, Listen, this is the story we want to show. So let's. Dig the numbers and cut them until we get this story. So I'm reminded of how my first career was in investment banking. And it was like, you can come up with any valuation you want. You just keep changing the inputs until the output is the right number. Have you ever been in that? Almost; I don't know what you call that environment.

ROMMY: Yeah, it's yes. You can use numbers to tell any story you want. We all know that. And it can be, it can be the story that's going to sell more products, right? It could be the story that's going to bring in more users. It could be the story that's going to bring in more VC money, right? It could be the story that convinces the board that your business is doing great. Yeah. There's marketing in every role, right? Whether you're an engineer or product manager or a finance person, it's always about how you take numbers and tell a story, and stories have many different flavors. And I absolutely, especially in the more corporate environments and the bigger companies that I worked at, and I've worked at a couple, will say that in those places, data is driving fewer of the decisions than what would be expected. There's tons of data available, and there's tons of data that leaders are looking at. But they're looking at it for, from a different lens and for a different reason, right? And I find that in bigger companies that I've worked in data was a marketing tool, marketing internally, marketing externally. Whereas in the smaller companies, we really want to be able to answer questions with data and we love to be able to, use data to make the most informed decision of how we move forward because we know that every day. Is another day we've survived, and another critical day for our business. So I think I agree, it's, I love data to drive impact rather than to do marketing. That's me personally. Some people don't operate that way. And some people really love to be able to weave wonderful stories with data. And I'm more like. Let me know where we need to drive the impact. And I'll find the analysis that's going to best inform that impact.

TIM: speaking of, storytelling and cherry picking data in a sense, that's what a resume is. It's a very controlled narrative about ourselves and our own skills. These days, which seems to be increasingly written or co written, let's say with Chatshippeetee do you still place value on a resume? Do you still look at it? If so, how do you think about resumes these days?

ROMMY: Yeah, it's a really interesting, it's a really interesting question. I think about my hiring process and I think about where it is now, what needs to change about it and what I look for it's, I'll give you a quick anecdote. We recently had somebody apply for a job and this person looked like the perfect candidate. And they hit. Every check Mark of what we were seeking for this role so much so that I went to their LinkedIn profile after getting their resume, they submitted, through, through the standard process on our website. And I went to their LinkedIn profile, and it was complete; it was completely different; nothing about it was anything that this person had said in their resume. And what I realized is, what I believe happened is they probably took the job description, popped it in a chat, GBT, and said, Build me a resume with these skills. And they submitted it that way because what they wanted was the phone call. Is the resume dead? No. I'm still looking at experiences specifically for roles where I require experience. I'm going to do the research to figure out, are you really doing the things that you said based on what I can find? And I have to triangulate things a lot more because of the challenges of, of kind of the world we live in. And especially in this tough job market, you have a ton of people submitting resumes and a ton of people who aren't even qualified for the job. And so that, that adds an additional challenge, right? So no, the resume is not done, but the work you have to do to read between the lines. is a lot more than we previously had to do, and because people are quite literally building systems to apply for jobs nowadays and using these systems to spam hundreds of different job postings a day, it's become really tough to sift through resumes. To me, I think. Resumes are one thing I'm going to seek people out personally on LinkedIn, who I think could be really qualified for the role. And so I think outreach to me is going to be even more important than inbound job postings. Now we still get most of our hires from inbounds, right? But you get a lot stronger candidates if you go and seek them out and do the sourcing yourself.

TIM: Yeah, it's such a tricky one, isn't it? Because these inbound job boards, in a sense, in theory, anyone can have a shot to apply. It's not some closed system. But because of that, inevitably the quality is low on average, to put it politely. And a lot of the applications are effectively spam. You would get the most random applicants; product analysts in Sydney, a chef in India has applied. Like I,

ROMMY: 100%.

TIM: Not going to get an interview. What's the point? I don't understand. And that must have just, yeah, increased so much since some of these automated application tools have come out. So yeah, going to a more outbound approach means, yeah, you've decided based on what I can see from this LinkedIn profile or whatever, this seems like the right candidate to get into the funnel. Does the hiring process then have to change as a result because suddenly it's flipped around and you're trying to sell them the job rather than they're selling themselves to you?

ROMMY: Yeah. There's also some fundamental bias that is associated with sourcing. I'm going to find somebody who appeals to me, and that's going to close the door to candidates who I might not consider or might not have considered. So I do think that you still need both outbound. And inbound, I think striking the balance between the two, I think inbound applications, that process is hard; that hiring, that recruiting, whatever that is, is completely broken. I think people are figuring out ways around the system to your point, like the chef from India or product analysts from Sydney or whatever that is. You've got teachers applying for data jobs, and people who are still, frankly, in the middle of college, applying for entry-level jobs. So to me, it's, to me, we have to figure that part out, and I think AI could help us do that. I think not quite yet. I think there's probably some bias that AI still has that needs to be solved. There's some hallucination that, you know, and I would need to be convinced that the way these. tools are evaluated, how they build evals is going to be critical to getting AI to the point where it can help screen some of these things. And maybe that's coming sooner rather than later. Maybe it even already exists, but I'm just not sure I fully trust it yet. But yeah, the hiring process is fundamentally changing because firstly, going back to something we said earlier, just how many data people exist out there now, right? It's not a world where You used to get a dozen submissions for a job. Now it's a world where you get thousands. And and, you used to be, you have to teach data people on the job because no one really had that skillset 20 years ago. And now people are coming in full fledged experts in SQL and Python and understanding data modeling. And, and understanding how to use data to tell stories and stuff like that. So it is a really different world. And so the hiring process is changing. I think. I think if I'm going to, if I'm going to seek out the best candidates, I'm going to have to do that myself. And I think that it's going to put more work on me, but in the end, I think it helps me to identify, what I'm looking for and helps me to also hone what it is I'm seeking in a candidate.

TIM: Is the fact that this falls on you as the hiring leader to do this a little bit tricky? Because in theory, a talent team is the one responsible for sourcing and screening candidates typically. But I found certainly in analytics over the past few years. Because it's maybe a niche area and it's changing so quickly that it's hard for talent teams to really grasp the, probably the subtle difference in their eyes between, I don't know, an analytics engineer and a data engineer or something like that. And so when they're just keyword matching, scanning, does this LinkedIn profile smell right? It's just not close enough, an algorithm that they have at their disposal because they're just not experts. Is that part of the issue, do you think?

ROMMY: Yeah, it is. Truth be told, I've worked with some truly. Incredible data recruiters. Folks that I would not hesitate to recommend to any company that have that do a really and, they, that's their niche. They focus on data recruiting specifically. So I think, I think recruiting in general is tough for the reasons you state, right? Recruiter hiring engineers is going to have a similar problem as a recruiting recruiter, hiring data people. But I think recruiting in itself is becoming a bit more specialized where it's not just, yes, some companies have just. Broad recruiters, right? We do as well. But there are entire companies dedicated to data recruiting. There are entire companies dedicated to engineering recruiting, right? And so I've seen the full range of recruiters some being incredibly impactful and just knowing exactly what you want and knowing exactly the questions to ask and how to and then, there's there are folks who are more generalists, right? Who won't get quite as deep? I think calibration is such an important part of that, right? So you get the first couple of candidates in, and then working with the recruit and partnering with the recruit is super important to say, This is why this person didn't quite hit the mark. This is why this person did hit the mark. And that level of calibration, as long as they're a really good recruiter who's able to adapt, is going to be. so important for the first, say, 10 candidates that you screen, right? So I think calibration is going to be a big part of it. I think, in some of my current roles, and mostly this is just because we're hiring a bunch right now. Hiring managers are doing the initial screen, and that's okay too, right? I think it's okay that you don't have recruiters jumping in and helping on that kind of stuff. But for me, I think hiring managers doing the initial screen also helps a lot. It's going to put, add more work to hiring managers, but you fight, you strike a balance with every job. And it also depends on the folks you have available and the time that you have and stuff. I think it works really well with recruiters, and it works really well with hiring managers, and it really works well with specialized recruiters. So I think that, in sourcing companies, right? To me, you just have to strike the balance for every role that you're doing and use the resources that you have at hand at the time that you're doing it.

TIM: You mentioned you're currently doing the initial screen. You mean you're screening all the resumes yourself, or you're doing that kind of first phone screen initial interview?

ROMMY: Both. Both. So we are, I think, hiring. I am currently doing my manager's hiring, but I'm jumping in to help out, especially as we get so many resumes and stuff like that. So I'll dedicate like half of a Friday, just going through resumes to help take stuff off their plate. But for me, it's both screening resumes and doing the actual 30-minute phone screen to start to make sure that they at least tick off the boxes on the basic things and are able to answer the basic questions that we're looking for enough so that we are ready to proceed them to the next step. So I think, ultimately. We're, I'm, I, myself and my managers, we're ready to take on everything from the resume screening to the to the first phone screen.

TIM: Have you noticed, I'm not sure if this is a metric that's tracked, but even if you have a feeling around it, have you noticed that the conversion rate from that resume screen to that first interview, if that has dropped over the past year, just because I feel like To your anecdote before around the candidate who'd optimized their resume to a job ad, and then it bore no resemblance to the LinkedIn profile, that must be happening more and more. So I could imagine how, yeah, the accuracy of the resume screen would be dropping.

ROMMY: Yeah, absolutely. So you know, I think the interviewing process is laborious for lots of people involved, right? You want a good cross sectional number of folks internally, you don't want to burn too much of your own time doing the initial phone screen or the zoom screen, right? You don't want to burn. Too much of your time on that and where you're doing like 10 or 15 a week. Cause that can get exhausting and you can burn out from it and then you'll be dropping other things and stuff. You want to be a bit more picky to who you move forward to the phone screen. But to your point, you're also going to be extremely picky from the phone screen to the next round of interviews. And the reason is I don't want to burn anybody's time. Who, with somebody who I'm. So about right. I'm only going to move the candidates forward that I think are a represented themselves really well to your point. And be, are truly the kinds of folks that we are looking for, whether culturally or technically, or, or just, curiosity wise, right? I think. I'm only going to move the folks forward that I really think have a chance. The other thing is I truly believe that everyone is entitled to go through the entire interview process. I don't like the idea of dropping a candidate mid interview process. I think everybody's entitled to the respect of being able to move all the way through the interview process. It's a, I think for me it's hugely important. So that means I'm just going to screen a little harder upfront right after the phone screen.

TIM: So let me see if I've understood. So you have the phone screen, which is, let's say, quite a high bar to pass. But then from that point forward, of which there are probably several interviews, everyone makes it to the end. So what you're saying? Oh, that's a really interesting way of doing it.

ROMMY: Yeah. Unless a role gets filled, obviously that's the only, that's the only exception, I think too, and it's, I can see the debate behind that, right? You don't want somebody to do a take-home exercise where they've worked so hard if you're probably not going to consider them, but usually if you do a good enough job upfront and you screen that person after that initial hiring screen, usually you want to see them all the way through because you want to make sure that if they're maybe slightly lighter in one area or slightly less experienced in one area, at least they possibly make up for it in a big way in another area. So to me, yes, I think everybody deserves the respect to be offered the entire interview process to prove themselves if they make it past the hiring screen.

TIM: Yeah, we were talking about candidates and seeing them through that whole. journey of interviews. And I guess that's because each of those interviews is evaluating different things. So then if you hadn't given them the opportunity, you wouldn't have the full holistic picture of their

ROMMY: Exactly right. And nobody's going to be, that's, very few people are perfect at everything, right? Sometimes, a weakness in one area might be made up for by a strength in another area. And, as long as you believe that those skills can be developed, then I think it's fair to allow them that opportunity.

TIM: And what that also does is help to solve a few other problems. One is just like the small sample size problem. Like you interview someone for an hour; one person does it. How much can you really know about them? Then you control for the different interviewers as well, because everyone comes at it from a different perspective. So you average out across all those little bits of randomness as well.

ROMMY: Exactly. It reduces bias in the process too. Everybody has different opinions on what they believe a strong candidate is. And so I believe that, like you said, you have to look at, holistically, the full spectrum of opinions on somebody, because diversity in the hiring process is extremely important. And so to me, having a diverse set of interviewers is as important as having a diverse set of interviewees or candidates. And it's only by having that level of diversity that you hire the best people to build a team with.

TIM: I like this approach also because I've often thought that hiring is almost the way it's set up in most companies as a one-strike-and-you're-out policy. Like you keep getting through the interview process. And as soon as you meet some interviewer who just doesn't like you that day, you're like, nah. You rejected it, which seems like a pretty harsh way of doing it to me, especially if that's a reasonably later stage. Like what the fact you've already nailed three interviews. Is that not something of value? Suddenly you get to this fourth one. They're like, nah, okay, but totally. holistic view.

ROMMY: Yeah, absolutely. And I think, interviewing is a very imperfect process, right? It's fraught with gaps. It's fraught with bias. You try to eliminate those things. You can't do it. So you have to rely on, as the data person would say, as the higher on the higher sample size on the higher volume of data, right? And so the more interviews that you have. Not saying that you should have a thousand interviews for every role, but like the more interviewers you have as part of that process, right? The more you're able to make an informed decision. So data is very much a part of the hiring and recruiting process as it is, in, in our day to day jobs.

TIM: Now, one thing as an almost devil's advocate I've seen is for. Let's say processes with several interviews and different interviewers. Yes, you get this kind of blend of opinions. But one thing I've seen where it sometimes falls down is, an interview in a certain stage, maybe isn't on the same page as everyone, they've got their own idea of what a great analyst is, that is sometimes contrary to what was in the job description, the job spec, what you're screening and searching for. They come along and go, Oh, this candidate didn't know. Ah, I asked them all these questions and they didn't know. Ah, so Paul. What are we looking for? Someone with our, we were looking for someone with Python as an example. So is it important to get people like almost on the same page, even with their different views from the get go

ROMMY: Yeah, absolutely. The best companies I've worked at when, when it came to hiring and recruiting at the end of the entire interview process, once a candidate has made it all the way through for every candidate, All the interviewers will get into a room because also what you put into greenhouse, your evaluation of the candidate, you're just trying to fill out the form and stuff like that. And maybe it might not be as thorough. I think it's important to get everybody you interviewed into a room to have the discussion and debate. Because what I find is occasionally you'll get those candidates where three or four people were like huge, strong, yes, big thumbs up. And then one person will be like, no, and what they say in greenhouse and stuff like that might not be enough. I think it's super important for everybody to get into a room and talk and go around and do a round robin and give everyone a chance to speak about what they liked about the candidate, where they thought the candidate might've missed the mark, and stuff like that. And they have that debate. And they use that debate to help the hiring manager and the HR person make the final call. I still think in the end, it's the hiring manager who ultimately should feel empowered to make the decision. But I believe that using the inputs of everybody is especially important and doing that on a live call. Is also especially important unless the person is an absolute shoo-in, right? Sometimes you do have a candidate that, like all six or seven people that are interviewed, all said strong. Yes. I'm not sure you need to get into a room for that one. I just don't think that there's enough negativity probably to have an informed discussion. You don't want to burn 30 minutes of anybody's time, I think, but I do like the idea of doing a live kind of round robin debate.

TIM: straight away in my head? I'm thinking of that classic American film with the 12 jurors, and one of the jurors that disagreed and 12 varying around exactly if you've had that situation where there was like one hard no versus six yeses or vice versa. And they managed to turn it around.

ROMMY: I don't think so successfully, but it doesn't mean no. It's usually where it gets fraught, like 50/50. I don't think I've ever seen a situation where one person came in and said, absolutely, we should hire this person. Everybody else was like, No way. And that person got hired. I would be fascinated, though, just to see that. I would love to sit in that room to see how that worked.

TIM: And then it comes down to back to the storytelling, your storytelling skills around conveying your view, your opinion with your data as to whether or not you should hire the candidate or not. It's all the same skill set, at the end of the day.

ROMMY: Yeah. Yeah, absolutely. I think the hiring process is a meta view of the actual job itself. And I think the best data folks to get in a room are data folks from all different. Backgrounds to share their opinions based on the different data points that they have to enrich the rest of the organization with the bank of knowledge that they bring. So with a candidate, I don't believe I hate the idea of groupthink; I hate the idea that everybody goes in, or everybody always has the same opinion in a company. I fundamentally disagree with that. And I've worked for companies where. You felt like everybody was always on the same page. And while that sounds brilliant in theory, it can be extremely detrimental to bringing in new ideas, new thoughts, innovation, debate. These are all critical things for a successful company. And so for me, diverse viewpoints are, is the richest data set you could possibly have. And the way everybody gets to tell their story sitting in a room of their interview with a candidate is also going to be critical to, hiring or not hiring that candidate. And no opinion is worth any less than any other opinion. And I don't care where in the organization you sit. To me. Every human has an equally weighted opinion when it comes to hiring someone, and I think that they should all be taken into consideration equally and the story that evolves from that is super important.

TIM: What about on the kind of technical skills versus cultural fit spectrum or what have you? How much value do you place in each? How do you view about these? Do you try to measure the cultural side of things? Do you try to measure the technical side of things?

ROMMY: we're doing it all I was gonna be such a cop out of an answer. But I'm like in the middle I. I, I actually lead more towards cultural fit being important, right? I think technical skills can be taught and they can be evolved, right? I, if you're hiring somebody who's super curious, if you're hiring somebody who has an eagerness to learn and to develop their technical skills, I will default to somebody with lower technical skills in favor of. Their curiosity, their eagerness to learn their problem solving mindset. That's so important to me, right? When I started off in data I learned on the job. I was hired onto a data team at eBay. I think we were like a 10 person analyst team at eBay at the time, which is super wild to think of. There were other analysts throughout the organization, but I showed up. I was the most junior person on the team. And I didn't know SQL at all. And, I think maybe for the first six months I wasn't even doing that. I was helping with process. I was helping with I, I was doing more coordination than anything, but being around data people was super intriguing. And then when they started throwing analyses my way, I was just, people were sitting down with me, teaching me SQL and stuff like that. And, where I was super successful wasn't in my technical skillset. It was. In my curiosity and my desire to, to understand the business and how the business worked. And back then eBay was the most, was the coolest business, right? It was a peer to peer marketplace. You had an economy that you were managing. You had, supply and demand. Like it was the most interesting business to analyze still to this day, probably one of the most interesting businesses I ever analyzed. And so to me, the thrill was in understanding them. The balance between buying and selling the economy, the dynamics at work, right? The economies of scale and all that kind of stuff are where I excelled. And then learning the sequel was the easy part, right? So to me, it's I'll default more heavily towards cultural fit, curiosity, desire to learn, desire to drive impact, and problem-solving skills. And people with different ways of doing those things is really important to write, like I don't want everybody to be solving problems the same way, or I want people who approach the problem top down versus others who do it bottom up, right? I want folks who. Maybe do some data discovery first to just see what the data looks like before they build their hypotheses. I want others to build their hypotheses first, right? Before they do, different ways, angles, and perspectives from which to solve problems are so important for teams, dynamics, and culture.

TIM: And so with cultural fit, then it's not necessarily about getting people who. Let me rephrase this. I've often thought that sounds like there could be a bit of a. Difficulty in having cultural fit and then diversity of thought that seems to me slightly oxymoronic. But they're not then if you are looking for, let's say, problem-solving skills, and then the diversity is the way with which you think about how to solve the problem.

ROMMY: Absolutely. So everyone should be a problem solver in data, right? How you solve that problem is where folks should differ. So for me, there are certain absolutes about what a data person should have. And it's just, it's the things I just mentioned around curiosity and problem solving and impact and business acumen. And storytelling. These are all the things that I consider to be. Absolutes for any data person, and that's whether you're on the kind of analytics engineering side of things with the engineering side of things or on the data science side of things. I think these skills, to me, might be weighed differently for each of those roles. But I think all those are critical to everybody. And, but a data scientist is going to approach a problem very differently from a data analyst, and a data analyst is going to approach the problem very differently from another data analyst. Everybody needs to have these skills, but how they employ these skills can be vastly different, frankly, should be vastly different. I learned from folks on my team. All the time in the way they solve problems in ways that I never would have thought of solving them. And as a result, I become stronger as a leader because I'm learning from them. So it's super important to have diverse ways of doing things on a team because it enriches the team, one plus one equals. At that point. And so it enriches the team overall. People learn from each other. People have healthy debates and people feel like they bring something to the table that somebody else doesn't. So to me, these are such important ways of recruiting and hiring a team that I think often go overlooked.

TIM: of learning from people, if you could ask our next guest one question about hiring, what would you ask them?

ROMMY: A question I would ask the next person that you interview here is what's your thought on hiring. Someone with no experience, someone either fresh out of college or pivoting from another role and teaching them on the job. And the reason why I asked that is I think one of the biggest inherent biases in the hiring process right now is the fact that so many companies look for people with data experience or look for people with college educations. And I think that's going to. Not be as, at least in the U. S. Not often be the case anymore. I think college has gotten too expensive here. I think that lots of stuff, especially data can be taught on the job. And so how comfortable would one feel hiring somebody with no data experience and teaching them on the job in an effort to a bring more diverse mindsets, right? Somebody with a completely, why not hire a chef or a teacher? What stops us from doing that? Because if we all chase after the same people with data skills, then it feels like we're not opening ourselves up enough to, to the possibility of bringing in completely new viewpoints.

TIM: I will level that question at them, and I'll also have a variation, which is someone is self-taught, so has no experience, hasn't done the official degree, but they've somehow taught themselves. I can think of some great engineers we've hired over the years, straight out of university, who had no experience. Experience, but they built and shipped an app onto the iOS store. It worked. It did some things like just doing that every day working here, and that's the job, so fine, you've proven you could do it. I don't need a four-year degree to see that you've got more in the bank, if anything.

ROMMY: One hundred percent learning through experience is probably the biggest and strongest signal of curiosity. That you could ever ask for, right? I was on a site called dune.com today, and I don't know if you're familiar with dune.com, but I would check it out. It's a site where you can essentially query APIs and query data on web3 applications out there, right? You can use SQL, and it's super straightforward. And I was on there playing earlier today. It's just unbelievable the analyses that people have built, right? And the data sets that people have built. And there's tons of sites like these out there, right? I just happened to be on that one today, which is why I brought it up. But, for me, some of the most brilliant people and the people you want to be working with are the people that figure things out themselves. We're self-taught. And I think that's a wildly underrated skill set to have is self-education.

TIM: I agree 100 percent, and yeah, hopefully the world moves more in that direction. I think it'd be fairer. Yeah, to not require people to have, as you say, increasingly extraordinarily expensive college or university qualifications that would open up things in a lot fairer way. So hopefully we move a little bit further in that direction. Rommy, it's been a great conversation today. Thank you so much for sharing all of your insights with us.

ROMMY: Thank you, Tim. Yeah. Awesome. Thank you so much.