Dev Interrupted

The Rising Demand for Soft Skills in Engineering | LinkedIn’s VPE Aarathi Vidyasagar

• Season 4 • Episode 34

Share episode

According to a LinkedIn study, engineers with strong soft skills are promoted 13% faster than those with only technical skills. Given how AI will continue to reshape market demands for developers and engineering leaders, how can you adapt to these changes and further your career?

This week we’re joined by Aarathi Vidyasagar, VP of Engineering at LinkedIn, to explore the growing importance of soft skills and how leadership, communication, and empathy are becoming just as critical as technical expertise.

Aarathi shares how LinkedIn is preparing engineers to thrive in this new environment, focusing on upskilling teams to navigate AI and empower collaboration and innovation through strong interpersonal skills. She offers valuable takeaways into building engineering teams that balance hard technical abilities with the soft skills needed to lead, mentor, and work effectively in diverse teams.

For anyone interested in the future of engineering and the rising demand for soft skills, this episode offers an important perspective on how to equip your teams for long-term success in an AI-driven world.

Episode Highlights:

  • 00:56 How industry changes have impacted the need for communication and soft skills 
  • 04:31 How software development is going to continue to change 
  • 12:33 Generalists vs Specialists 
  • 15:52 How is LinkedIn positioned for the new era of hiring?
  • 31:11 How is LinkedIn approaching developer experience and productivity?
  • 36:33 How to support devs in automating migrations 

Show Notes:


Support the show:

Offers:

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Soft skills have been something that's important throughout, right? I mean, it's not new, but with the onset of generative AI and coming in and taking away some of the mundane, how do you differentiate yourself? So as an engineer, you, you've got to bring things that are not, you know, Easily applicable. And so things like communication, leadership, soft skills become even more important. They've always been important. Like if you look at, you know, success criteria, if you look at leaders in the industry today, you would see that a lot of it has to do with communication, leadership, like a soft skills do play a big big role in terms of growth, etc. some of the other skills that kind of helped you stand out might be easily replicable now with generative AI and the tech changes that are coming in.

13% of all pull requests are bot created today and they are creating a unique impact on your SDLC. LinearB his upcoming research exposes the effect bots are having on our teams developer experience and productivity, and engineering org who created a system for managing bot generated pRS are able to reduce their entire review load by over 6% while also making drastic improvements in their security and compliance posture you want to learn how your team can manage bot generated PRS and get early access to linear B's data report. Head to the show notes to register for our upcoming workshop on September 24th or 25th.

Conor Bronsdon:

Hey everyone. Welcome back to Dev Interrupted. I'm your host, Conor Bronsdon. And today I'm delighted to be joined by Arthi Vidyasagar, VP of engineering at LinkedIn. Arthi, thank you so much for joining us on the show today.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Thank you for having me. Excited to have this conversation.

Conor Bronsdon:

You've got some really interesting perspectives that you're bringing. And according to new data, Recently released by LinkedIn, the demand for soft skills among engineers is on the rise. Uh, it's something you and I've talked about a little bit in this six month period from 2022 to 2023 across the US, UK and India markets, communication was consistently in the top five for skills, getting the most share and demand. And at the same time, I know there's a previous LinkedIn study that found that engineers with strong soft skills are promoted 13 percent faster than those with only technical skills. And we've seen this, even continue to rise in importance as this is happening with the backdrop of Gen AI's transformation of software development. Do you think that these industry changes are helping drive the desire for communication and other soft skills among engineers?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

I think I want to start with community soft skills have been something that's important throughout, right? I mean, it's not new, but with the onset of generative AI and coming in and taking away some of the mundane, like, you know, your coding, et cetera, it's, it's how do you differentiate yourself? So as an engineer, you, you've got to bring things that are not, you know, Easily applicable. And so things like communication, leadership, soft skills become even more important. They've always been important. Like if you look at, you know, success criteria, if you look at leaders in the industry today, you would see that a lot of it has to do with communication, leadership, like a soft skills do play a big big role in terms of growth, etc. But now it's just ever more important, like, you know, going back to like our mathematical theorem, while it was sufficient, once upon a time, I think it becomes necessary now with the changes that are coming in, because some of the other skills that kind of helped you stand out might be easily replicable now with generative AI and the tech changes that are coming in.

Conor Bronsdon:

And we've had variations of this conversation here on Dev Interrupted before, our listeners may be kind of groaning hearing me talk about communication. They're like, Oh, every leader who's really good at their job talks about the importance of communication. I think the difference here is LinkedIn. has this massive pool of data to back this up and say, look, we can actually look at the job market and say, this is something that's highly valued. This is something that is increasing in importance, and it aligns to your point to these transformational shifts that are happening in how software development works today.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

I think, look, LinkedIn's been doing this whole skills based hiring, understanding skills, mapping skills. Like, this has been a journey that's not started just because of the generative AI movement. But now that it's happened, us having this information and this breakdown just makes it like even more important to kind of see where the trends go, right? Like what are the trends that people are looking at? What skills are like most important here? How do we begin to differentiate? This is where we as engineering leaders have to come together, start looking at what we now call this talent map or skills map to see what those gaps are, what those trends are and where do we need to go. So I think it's super important to know. That soft skills, again, not easily replicable. So again, nothing new, right? This is not a new formula. This is something that's been true. Uh, for us to be good engineers, you've got to bring the set of people along with you. Communication becomes key. Again, we're not talking about oratorial skills, right? That's not what we're asking for. Just being able to like communicate your ideas, bringing the team along is, is what you're looking for. And a lot of that distills into what differentiates. It's that. A little bit of salt in the dish that makes you better.

Conor Bronsdon:

And I think you're really apt here because we see other technology transformations that may not necessarily wholesale change what's happening within a market or industry like software development, but it does accelerate underlying conditions. And, uh, As you pointed out, communication has been crucial for leaders for years. How do you manage a product team? How do you make sure that your relationships within the org are tight so that you get the backing and support and resources you need? How do you make sure your team's actually staying on track and aligned to business priorities, actually getting the work done that you need to get done? But we're also seeing this major transformation in how software development itself may work in the coming years. Whether that's on the no code side, whether that's prompt engineering, changing where we spend our focus on workflows. How and how much do you see software development changing in the coming years?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Uh, I think again, going back to us understanding skills. I think there was a study on LinkedIn, uh, where we saw, or we were expecting that 50 percent of the skills to do a job, it was going to change by 2030. And what we saw is from 2015 to 2023 is when. That kind of change happened. With the onset of, you know, technological changes, like in a generative AI, we expect the curve for that to become a lot steeper. As in, like, I think in another five years, we will be expecting, I think we expect it to be 68 percent of the skills to do a job is going to be changing. Like, to put it in perspective, two thirds of what I needed to do my job You know, 15 years ago, it looks very different. So there is a saying that our CEO uses quite a bit, which is jobs are not change. I mean, we're not changing jobs, but jobs are changing on us. And that's very true, right? So what skills we need to do the same job? 15 years ago to today is going to be completely different. So with that as a backdrop, so really think about where we are as software engineering leaders, just software engineers, right? So we have to be upscaling, reskilling, thinking about what does it mean? What does it mean to be a full stack engineer is very different from what it was when I started life. And now I'm going to begin to date myself and tell you I'm a fossil literally, uh, to, you know, what it is today. I mean, there were like. Differentiated fields like, you know, ops engineers or SREs, those quality engineers, different stuff. And over time, all of those have been merging, right? So the team structure, the team formations have changed. And with the most recent sort of changes with AI kind of becoming ubiquitous, what we see now is I think full stack should include A reasonable awareness, understanding and competency in, in, in the field of AI, ML, and, and, you know, understanding data science. Like without that, you can't be an effective software engineer. So how do you then, as a software leader, kind of help enable, you know, transform them? Your team's that the right way. Again, nothing that we haven't done before. Remember, we've gone through the new QA revolution. We've gone through the, you know, ops engineering, becoming DevOps. Um, so it's not like this particular ecosystem has sustained all these changes. We've been able to do it, but I think it's a constant change. The big differentiation now is that the pace of change, I think, is what is going to be the challenge that we'll have to keep up with.

Conor Bronsdon:

It is funny to think about this because I think you'll hear a lot of successful people like yourself talk about this need for reskilling, this need for being a lifelong learner and loving learning. And my wife, Catherine, talks about the fact that her job that she does, which is CRM automation and lifecycle management, didn't really exist when she went to high school. Like it was barely a thing. Salesforce was just getting started. and this is, I think, true for a lot of us where the definition of, as you point out, a full stack engineer has changed over the years, uh, how we do our jobs day by day is changing over the years. So how are you approaching that skill transition at LinkedIn for your team of more than 600 engineers?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Again, like I said, I think the challenge like It's not a new problem. Like, we've had to reskill, upskill changes. I mean, again, at the risk of dating myself, when I started, like, when I studied CS, um, it was more in the Perl C And I don't know, for folks that are like, you know, developers, like, doing memory allocations, knowing how, like, sort of the basics work to Java to where we are today. Um, I think a lot has changed. I think the big piece here is the pace of change, the rate of change that is going to be there. So I think this is where being, that mapping of what skills we need to, where we need to go. So constant need for upskill, reskill. Constant need to sort of think about what are some specialist roles and what are some generalist roles in the team. So as, as leaders, you know, A, encouraging. I mean, like, look, most. Developers, most people like me that have gone through the cycle, are honestly academicians at heart. Like, you know, this learning and getting to learn, it's just, it's second nature to us. You've gone through so many years of studying, you know, this whole learning is something that comes more and more naturally. So I would be like, really highly recommend that. But I think with the Pace of change. One of the things that we are trying to do, and remember, I am, as part of my job, I also run the LinkedIn Learning product. So this is where, like, you know, curating the right bite sized learning opportunities to upscale. So there's this, this Like, how can I give people the right amount of knowledge so they can get to the right level? Because I don't think we're getting to PhD level of upskilling, reskilling. It's going to happen at a rapid pace versus can I know enough to be That sort of, you know, reasonably good at being more of the generalist. And then of course, there's going to be the need for the specialists and then opportunities to get deep, deep in there. And that's a whole different, like your traditional education system. But a lot of what we're trying to do, even with LinkedIn learning courses, and here's where I'm going to make a pitch for my product is putting those bite sized, you know, learning. So we've, I think, generated over 250, like, AI based courses. We do about 14, like, 23 different languages. This is where, like, I think, and this is not just LinkedIn Learning, right? Like, most learning platforms now are coming into that world of giving the right, like, sort of set of information that's current, like, tasty enough. So just to give you an example, some of the things that we've tried with, with my sort of team is, A, educating the leaders. which is the managers, to become savvy and aware to understand what generative AI means and what does the technology stack mean, what do systems, what's the machinery that goes into it. And then sort of enabling much deeper coursework for more hands on, you know, giving opportunities for learning, for engineers to kind of, like from an apps domain, sort of dabble into prompt engineering, which is probably the entry level into more of the data world. So I think just sort of encouraging that slow transformation, again, to get more of the data generalists in, in the system as, as the full stack, and then sort of giving the right opportunities for people who might want to go deeper into that specialist world. I mean, there's a big difference that the deep AI folks are the PhDs who understand all of the things. Depth of, you know, deep linear regression equations, et cetera, which is not something that'll come naturally. I mean, I, for example, hate statistics and never survived or thrived in it. but again, need to know enough to be at least aware. And empathize with the team and and be able to do that. So again, a big part about is is giving encouraging that. So as a culture, sort of changing the emphasis for learning and reskilling, providing the right opportunities for people to get that taste of, you know, what that is. Reskilling would look like to, you know, deeper and better opportunities there. But again, I think this is where the opportunity for online classes, democratizing some of that education system, I think it's going to become more and more important for us. Again, technical revolution here. It's easy to curate and make those courses, and bring it to life. I think you see a lot of startups come in here. And as a learning product owner, we see a lot of that competition as well.

Conor Bronsdon:

You mentioned something there that I want to dig into a bit, which is that there's definitely a need for improvement for generalists, people to pick up skills and understand some of these new technologies, some of these new approaches. And there's also this need for specialists. And I wonder if you think the ratio of those needs is starting to change. With, uh, the current era of software development and how it's transitioning. Do you think engineering leaders should be constructing their teams or changing those ratios of specialists to more generalists, uh, or, or maybe even like more leadership roles, uh, compared to how they were doing it in the past?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Um, yeah, I think, like, remember, there was in the previous world, or at least the traditional sort of differentiation between AI versus apps, if you really think about it, the AI teams were mostly the specialists that went deep into sort of the models, etc. I think with this sort of change of tech and the pace at which we're going, we're going to get to a world where a lot of our sort of Platforms, open source, going to be the no code, low code kind of areas where lots of the model marketplace is going to be just like robust, right? And, and then there's a little bit of the marriage of what you want to do. So, so I think this is where You can be like pretty effective in being able to sort of use that catalog of information with the sort of generalists. So you do need the generalists, but you can't like, I think that the need for specialists doesn't go away because who then builds these platforms and the like the deeper sort of, you know, the chat GPT kind of technology platforms, right? I think it's just, it may not be the nature. I think there's going to be a whole segment of uh, you know, platform development that changes to this whole, how do I build the low code, no code, you know, broader case of depth there, and then the utilization of that and how that sort of changes into the applications and the applicability, I think is going to become different. So I think there's going to be a little bit of difference. So for somebody like me that runs like an application team, I that builds on top of a lot of these platforms, I would expect that I need to get more of the generalist while start beginning to depend on more and more of the platform developers. Um, I mean, like our partnership with OpenAI, Microsoft kind of stuff, like more and more. We see the pace of change, the depth of sort of changes and the technological changes, the pace at which it's going. If you're not somebody who's innovating in that sort of deep space, you will not be moving at the pace, like it'll be hard for you to kind of keep pace, you'll keep yourself, you'll keep falling behind. So I think this need for adaptability to this open source, being able to sort of, like, at the current pace, I think it's going to become more and more important. So again, it depends on who you are, like if you are, you know, AWS, OpenAI, I think, The, the, the, the kind of mix that you're going to have probably your need for specialists more, versus if you were a Salesforce or Workday or a product team or a startup coming up in an area, I think the kind of things that you need to do would look a little bit different.

Conor Bronsdon:

That makes total sense. And I imagine this is somewhere where LinkedIn probably feels well positioned, given that we're seeing more and more repeated tasks become automated. And LinkedIn's had this consistent skills first hiring approach. Do you think that LinkedIn as a company is well positioned for this new era?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Yeah, I mean, just think about the mission for LinkedIn, right? This to connect every member of the workforce with opportunity. And so this has been a problem space, like work and connecting people to work has been an area that we've been ideating for, you know, years now. This is what we do. Um, so, so again, having that framework of skills just kind of helps us like bring and automate, make that automation that much more easier. Right? So even for us to be able to sort of think about what are some disruptable skills versus what are some augmentable skills with generative AI? What are skills, you know, that new skills that might come up to where, where is directionally, where are people going? I think this is where LinkedIn's well positioned a because of who we are in the platform and our sort of economic graph, understanding some of those changes in the nuances and also helping that transformation.

Conor Bronsdon:

Right.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

As we help, like, you know, define jobs, understand jobs, work, uh, I think being at the heart of that is going to be a big piece of what we do. And that's actually why I come into work every day and I'm excited about it.

Conor Bronsdon:

Well, you're making a good pitch for the company. So, uh, I'm, I'm sure the PR team's happy to hear it. And I do think though, there are going to be some engineers who listen to this conversation and get a little worried and think, you know what? How do I balance sharpening my technical skills, especially as the kind of what is state of the art is constantly changing and changing aggressively right now, while also trying to develop, you know, more leadership skills and move into leadership and add my communication skills? Like, how, how do you balance that as someone who is maybe early or mid career?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

I think my advice to people is like, you know, this always on learning is a, is a really, really good mantra over somebody who's in any kind of like a technological role. Um, I, I don't think, um, it's, it's, it's a new thing, right? I mean, if, if every leadership conversation I've had, I've talked about, you know, I've always been a lifetime learner, challenges, like, you know, sort of pushing myself on the boundaries, um, of learning new things, not becoming too comfortable has always been something that's driven me. Only difference now is, again, I go back to the analogy of it was a sufficient thing, but now it just becomes necessary. So you just have to become very intentional about it. And I don't think you can, like, Just leave it to chance or be organic about it, right? So I think this is where you also have to understand that things are changing. You have to keep pace So you kind of have to invest in that learning cycle just becoming very intentional and this is not just on the engineers and the individuals So I don't want to throw the onus of that, you know upskilling, reskilling Just on the engineers themselves. I think we all as organizations, need to a, enable that transformation, recognize the cost from that transformation, um, and, and, and shepherd that like, and if we don't, if we can't do that, I, I think we will fail our engineers and, and, and that's not the right place to be in. So I think as a leader, I kind of see that as, one of my biggest roles in this moment is to make sure, how do I make sure. I give people a space to learn. I make it okay to spend that time and, and just make that a very intentional thing. I mean, again, just a little bit of the culture thing. I think long time ago, I'd read a book, Atomic Habits, so change little things, you know, and it sort of becomes a pattern over time. So this bite sized learning kind of fits into that. Like, you know, how do you then encourage that, recognize that, you know, give people the space and time. So I think this is all very. If we as organizations need to start making this kind of move from that sufficient condition to necessary condition transition just needs to be something that we all have to come together and make it very intentionally.

Conor Bronsdon:

I completely agree. I think. It can be challenging sometimes, especially in high pressure roles at companies that are driven to achieve to make that time for learning. And that's where I think leadership and organizations need to be really intentional because, uh, it's hard in the moment to see, Oh, well, I have to deliver, you know, these different epics. We need to complete this new product line. We have these challenges to do. Uh, we have products to roll out. We have, you know, business critical things. But if we ignore the bigger picture of how are we building for the long term, how are we building our teams for the long term, we burn people out. We don't set them up for success at the next stage and, uh, we risk so much. Um, and that could be a really challenging thing for leaders to navigate. So I'm, I'm glad we're having this conversation about it. Um, I want to drill back in around soft skills, because we've talked a bit about communication and some of these kind of crucial project leadership, project management skills that, really help enable people to get things done at companies. What specific skills or techniques would you advise people who are moving into engineering leadership to think about honing?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Yeah, I think see, communications leadership sometimes becomes such generic terms. for me, I think, look, end of the day, communication as a leader, sort of the biggest challenge you have to, like, face, or you will face is bringing your team along. on ideas, product, technology,

Conor Bronsdon:

Almost like evangelizing that part of

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

And, and, and so the most crucial skill, which then as you grow, you know, through those sort of corporate ranks becomes like, that's a skill that you start using outward, like to your, like, Peers, then to your sort of, you know, superiors and then like, you know, to, to cross-functional leaders. Uh, just the, so that the, the scope or the approach for that kind of changes. So I think a biggest piece is that that crisp communication of. Like intent and, and the why, and, and the idea of bringing people along. Because it's never, I mean, a lot of people, I mean, I've got teenage kids who are like, well, you're the boss. You get to tell people what to do and they'll do it. And in, in all honesty, that's not how the world works, right? It's a little bit of, I've got to convince my team. This is what, you know, this is a great idea. Here's why it's going to be good for us. This is the vision. You got to get them excited because when you do it right, it's a wonderful experience versus like, you know, do it because I say, so it's just never the right sort of form to do it. So I think for, for me, communication is about that. Like, how do you sort of message and appeal to the, you know, the, your audience? Whoever that audience is, whether it's your boss, whether it's your peers, whether it's your customer, or even just your team, I think just distilling that, and that's a, and it is truly soft skill, right? Because it's a little bit of a, it's more of an art than a science at this moment. It's like understanding your audience, what, what sort of makes them tick, like in that moment, sort of distilling that, and then being able to communicate. So I think it's just that, and it's also why Something, you know, that, you know, AI probably can't sort of do it as effectively because you and I are talking. So this is me understanding what is Conor like, or what would sort of, you know, make him tick and then sort of, you know, adjusting it. So it's also like maybe in some way saying, There is a salesman role in you that you have to hone and sort of bring it to the right sort of efficacy. So in all honesty, sometimes I've always joked about it, but I do think it's a, it's a, it's actually a good exercise is everybody should do a tour of duty in sales. Is the thing that'll help you kind of like learn some of that, that art, because that's the art, right. From a communications perspective to kind of nail it.

Conor Bronsdon:

I hate to admit it, but one of the most formative jobs I ever had was like a summer job when I was in college where I did door to door fundraising for a nonprofit. And it was really, really hard. It was like honing sales skills, was talking to all these people, was trying to, you know, develop my emotional intelligence. I I cried a couple of times doing that job. It was hard. I had like quotas to meet. and I'd also, it's one of the things that I've learned the most from and I've been able to like grow as a person out of. So, uh, I completely agree. I think this kind of ability to sell an idea and pitch people, like I think a lot of folks in software development want to think they're I'll say like too good for sales sometimes where it's like, Oh, my product's going to be incredible, it's going to sell itself. And there's something to be said for product virality and, and. The ability to have a product that can, you know, have a great K factor and spread itself. But, uh, you still need to be able to be great at distribution. You still need to be able to get your team on board because to, as you said, the difference between someone who's inspired and fired up to work on something and thinks they're making a difference and thinks what they do is impactful versus someone who's just checking a box to Get a salary is huge. it also brings up something else, which is kind of the collaboration factor. So you've talked a lot about like, how do you inspire a team? How do you bring them along? How do you get your team to work together themselves better?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

again, a lot of software engineers, developers that I meet, like, I mean, all of us engineers, we like to leave, we are introverts, which we probably are, by the way, right? And so there's this, there's a myth of It's not something that comes naturally to you. But I think they are actually pretty orthogonal, right? I mean, yes, you can be an introvert. That doesn't mean you cannot communicate or learn that skill and be able to, like, you don't have to be a social bee to, to be able to do what this is. I honestly think I'm, I'm, I'm an introvert, but I mean, here I am talking to you on a podcast and, and, you know, and I guess talking more than one sentence. So in the moment, it's not about that, right? It's, it's more about what you feel passionately about. And can you bring that out? And, and doing justice to who you are, I think, becomes an important thing. Again, like, it's, it's those things that, like, don't convince yourself that you can't, or you won't be able to do it, because I think they're not, they're two separate ideas and concepts that you need to keep. Again, very, very important, like, for us to be able to do this. I think people stumble through it. Like, when I was younger, or when I was contemplating, like, management, um, I always believed that, you know, I won't be good at it. Like, you know, I just, I, you know, I'm, I'm a developer. I love, you know, doing me, my computer, my program. And that was like what it was, that was who I was. And somehow I didn't believe that I would be able to, you know, bring people along and lead. And that somehow scared me. Um, but everybody around me was like, You're really good at it. And I was grateful for having that sort of, you know, peer group that encouraged me to do that. And I've been a leader, um, for like over 20 years now and, and I have never looked back. But when I originally took over the role, I remember like talking to our director then sort of making sure it's a completely reversible reaction. And, you know, I can go back if I didn't like it. And, and she, she was. She just laughed at me and said, I don't think you're going to need it. But there was something about it. Again, this is where I would say, I think very, very important. Again, not like that. There's a lot of myths and things that we say to ourselves, which is not true.

Conor Bronsdon:

Yeah, I think there are a lot of folks who are in leadership who start off If not continue to have this self doubt, I know I do. and for me, I know it's part of being very self critical and thinking about like, Oh, well, like, yes, I'm good at this thing, but I have so much to improve. Like I see all these holes. I see all the gaps. I, you know, I, I seek that critical feedback. So I, I cannot see how much I still have to learn. And I think it's really easy to let that self doubt, unless you qualitative or quantitative. To kind of rule you and let you go, Oh, well, I, I'm not the right person for this. And as someone who also is like a natural introvert and, you know, recharges from, you know, sitting at home playing video games and working on my own products and reading a book, like. Building those extroversion skills, those external communication skills has had to, at times, be really intentional. Um, and I think something I found is that over time I find that I, I, I gain a lot more joy from doing those tasks, doing that kind of communicative organizing, but it's something that, uh, it's a passion I've built, not something that I kind of came hardwired with. And I suspect you maybe feel similarly.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Yeah, I can guarantee you if you pull a sample of engineers, that will be the answer. But again, going back to you have to be intentional on some of these. Barrier to entry exists, but it's mostly in terms of how we've convinced ourselves. I think they are very doable. Um, So, yes, and it's just going to become even more critical as the landscape of work is changing around us and what we bring to the table. So, again, very important for all of us to kind of push the boundaries on what we can, right? Um, I think that's sort of the key. So algorithmic SMARTS is going to be so automated that that's not going to differentiate you. So you've got to figure out what that differentiation is going to be. Again, no different from, you know, any transformation that we've been through. That's the piece I want everybody to anchor on. This is the conversation I have with all of my engineers as well, is this is not the first time we're making such a transformation. This muscle exists. We've, we've improved on it. Uh, I think it's just a matter of continuation. It's just the tactics and the pace is what differs in the moment.

Conor Bronsdon:

And I'll say tactically, I love that when you first took on a leadership role with an engineering, you were given that opportunity to try it out too. I've talked to several folks who have had that experience. We at LinearB have had people, you know, try out being a team lead for a while and say, you know, I, I actually do want to be a principal engineer. And so I really want to take that technical track. Um, and so I think this is where companies that. And mentors that kind of enable people to try on leadership and learn is so crucial. And so I love that you're bringing that perspective to your teams and really encouraging micro learning and building leaders throughout the organization, uh, to, to lead from every seat, to borrow a phrase from some, some folks. And I know there's another concept, uh, that you've thought quite a bit about, which is developer experience. Uh, you know, how can we help our developers, help our team leads, help our entire engineering organization, help the industry itself. Be a place where we are more joyful in work, where we, get to deal with less really annoying, minor headaches that make our jobs harder and get to be more inspired about the products we're working on, which is something I know you're passionate about. And this obviously ties in tightly to developer productivity. They're kind of, twin concepts, almost two sides of the same coin, and it's becoming increasingly important, increasingly talked about, with the rise of AI assisted code, this trend is only continuing. How are you approaching these twin concepts of developer experience and productivity interval?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

I mean, LinkedIn has always been in the forefront of developer productivity, uh, and experience. Uh, just, just anecdotally, um, you'll see I'll kind of bring some of these stories from sort of experience as well. Um, I moved from my previous company to LinkedIn, um, I joined as a senior director here. Um, and the first thing in my previous company, I'd spent like 19 years. So I had like written code growing up there. So I knew practically every semi colon in the code base. So coming into a company like LinkedIn, not knowing all of the code was something that I felt kind of, you know, untethered in the moment. So, and I, I remember like asking my teams like, Hey, help me sort of write code. Um, I want to say that in my first week, I checked in code. I think it was like 17 minutes from, you know, time to check in to actually have, you know, my two lines of if then else statement, however small it may be, to make it to production. So it sort of makes That production footprint, this was six years ago, was like phenomenal sort of to get there. So, so LinkedIn, like to some degree has always sort of, you know, prided itself on, on being that company that's at the forefront of thinking about, you know, the like. Develop, Reproductivity, Happiness, and going first class at it. But the way, I mean, just bringing us back to the conversation, I think there's also a moment with all of the technological changes with generative AI, etc. The opportunity for disruption here is high. So, so we're going to see a whole bunch here where, whether it's within the company or whether it's third party tools, where we're going to see more and more of that efficiency sort of coming in. Right. So, so again, in terms of how are we thinking about it is a, of course, you know, bringing in that disruption, bringing that tools, being aware of what those efficiencies are. I mean, you know, we are in a mature company with code bases that are over 10 years, obviously, migrations and things like that. Typically, there's always the step. Tax, what you call it, like, you know, the sort of tax of maintaining a, you know, legacy code base. I think making all of that cheaper, better, easier becomes something that, that we can, like, actually improve on. It's not one of those things that everybody's like, you've got to do your time is no longer going to be true. and so excited for that, but I kind of anchor it as back to, and I believe this is also some of the, Developer productivity, happiness, sort of a framework that we've talked about is linked in in the open source world is more on like measurement, like it starts with understanding where your bottlenecks are, measuring the right stuff, um, and then sort of cranking down on where those inefficiencies are in the right tools, uh, whether it's building the right tools or building Buying the right tools, I think, given the market and where we are, I think a lot of that is going to change in terms of landscape and sort of, so step one would be to know what those bottlenecks are. I think most recently we've done a few of our sort of API migrations, which have been like, Pretty much, you know, code gets generated, sort of put into the system, runs the test, like very few breakages where human beings get involved. again, I will challenge you to find any single developer who will probably tell you that's a bad idea, or I'm afraid that my job's going away, because it is not, because this just gives up more time for you. But, A, it's a few things as a team, as an organization, is A, giving that measurement, understanding your ecosystem and where your teams are spending time and where they, what is the mundane time versus not, like just instrumentation, and maybe that's the right way to sort of frame it, to bottlenecks and then sort of bringing the right sort of tech and tool, um, and enabling this. But this is one area I think which would be like, it's the right marriage, if you will. Right? Like, I think this is where it'll be like, I don't think you'll find anybody that complains about it. a lot of the sort of agentic world or, you know, agents, etc. is what we're talking about in terms of generative AI becomes this, you know, somebody's doing the work while you're like not actively involved. That's huge. Really think about it. Sometimes I jokingly call them smart cron jobs. But at the day, this is again, me dating myself, but at the end of the day, think of the power of that,

Conor Bronsdon:

Yeah. I'll say we're, we're putting out some research soon. It might actually be out by the time this podcast comes out. We'll see about what we're seeing around the data on AI agents and automating PR as an automating, uh, code and increase that velocity. I'm, I'm excited to share that with the world and for folks who are looking to kind of start from those base principles you mentioned of like, okay, we want to understand our developer experience and our productivity. Um, you can go to LinearB. io slash resources. We can get your met Dora metrics to you for free. Uh, and there's also a guide to kind of setting up software engineering intelligence so you can have that observability of what your processes are like, you can set up a free survey instance and kind of get rolling. And there's a lot more you can do in depth there, but that's, that's a great place to get started. Um, you mentioned something really interesting on the automation side of things and kind of approaching automations to support your devs about automating migrations. Can you expand on that a bit?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Again, I'll anchor back to the sort of easiest example, right? I mean, when, when you're a large company, you've been around for like, you know, 20 plus years, You've got a legacy code base and, you know, you change code. Like sometimes front end technology changes, you know, that's also rapidly changing in terms of back end, in terms of how we do APIs, containerization. There's a whole bunch of, there's a gamut of things that you kind of have to upkeep. to to sort of adapt to the newer world. and he says, again, think of what we typically did to kind of make through some of those changes. Um, I think this is where, uh, previously it would be a set of things that we'd have to do libraries that you have And then you kind of, you know, upgrade all your builds, there's a huge change list of things, some are not backward compatible, so a human being has to get involved, and you kind of put this in, like run through some manual testing or automated testing, sort of go through your sort of certification process, etc. Imagine a world where Like, your biggest bottleneck for why did human beings have to get involved is that whole sort of resolution of, I can't, it's not backup compatible, I need to change it, like, da da da. So all of that, if, if in today's world, if you can understand better, do better and sort of change it. I think that that removes that barrier to sort of, it makes it that low touch. And then so you build it in a way, yeah. Where we're not spending hours and days, like, you know, it becomes grunt work. I mean, I think accessibility changes sometimes become like that too, right? Like, you know, you kind of have to get the right sort of form factor in there. Sort of your scrolls, et cetera. Sometimes it is very cookie cutter work. It really doesn't require a lot of like, it's just standardized changes. Um, you could get a lot of that. Going faster, better, uh, and easier because it's no longer just like patterns, but it's smart patterns, right? This is where you've understood and you've seen enough code that you know how this is resolved. Again, it's for a company like McDonald's, got like, you know, millions, you know, terrible millions with, you know, lines of code here. Um, easier to kind of get some of that understanding and the smarts in there. So again, this is, opportunity is large.

Conor Bronsdon:

And I think particularly for enterprises like LinkedIn, when you have these aggregate gains where maybe you save a developer an hour every week, uh, and you see that go across the organization, that is a massive savings. One area where we're already seeing that is JNAI is being leveraged to write tests for code, uh, and that's making an immediate impact. Uh, it appears there may be an even bigger opportunity ahead here. What are you seeing in that area?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

and again, again, I want to anchor on yes, time saved is great for the company because you sort of increase your productivity, but you're gonna take away from the happy developer, right? I mean, this is like, I cannot put a A value to that because I can't tell you how many times I've had, like, you know, developers come to me and like to have career conversations. It's like, almost feels like I'm just kind of doing this rinse and repeat stuff. I'm not learning much. I'm not innovating. You know, so there's always that conversation of, you know, I'm not I'm excited for doing all of this stuff. I'd like to build stuff from, you know, scratch. Like, what is the lure of, like, startup sometimes? It is that, right? Like, I can actually, I don't have the weight of a larger code base, and I can just, like, build ground up stuff to, like, now you Can actually give them the time to do some of that. And, and it's exciting and you kind of have the best of both worlds. So I think this is where I would say, yes, it's time saved and productivity for the company and probably, you know, more code and better code for us. But at the same time, it's happy developers. So it's just, and then motivated developers. So, um, I think that's something that you can't take away from it.

Conor Bronsdon:

Do you see this opportunity to improve other day to day processes like code reviews, for example?

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Yeah, code reviews, bug triage, um, you know, a lot of LinkedIn, you can do a shake for feedback if you don't like it, and the shake for feedback comes in, um, it goes through some level of automation, but eventually there is a set of human beings as you look at it, imagine a world where you could do a shake for feedback, goes into a system, the triage system is able to pick it up, replicate it, debug it, to a level and then sort of, you know, bring it to the, finally, when it does come to the developer, it's also literally, like, fully, here's the, you know, stack call tree for where the problem's happening or not happening, kind of stuff. So yes, I think there's this opportunity here in terms of, like, areas of software development that can change. Again, every time I say this, everybody's like, does this mean I'm not going to have a job? It's like, and I keep saying like, that's not true, right? I think the kind of job that you're going to do is going to be looking different. Going back to the point I had made earlier, the skills you needed to do a job to that's going to look very different in a few years. It's just what you're going to be doing in your job probably looks very different. And hopefully you're having fun doing that, and that's how you should take a look at it. So again, in the moment, that would be the advice I would give to people.

Conor Bronsdon:

I think you're spot on. And I think it's really true of almost any job, which is if you stay long enough. You will have to learn things. You will have to learn a new database. You will have to figure out these new skills. You will have to grow to stay in that role because the company's going to change around you. The goals are going to change around you. The world's going to change around you. That's like the constant of life. And, uh, I think what folks are maybe nervous about right now is the pace of the change happening, how transformational it is in some of the areas where people have felt really confident on the technical side. They're now going, Oh, this is really changing. But if you develop your skills here, if you do these micro learnings that Arthi's talking about, and if you've built your communication skills and work better with your colleagues, uh, you're going to be really well positioned for the future here. So Arthi, thank you so much. I really appreciate you sharing your insights with our audience and I hope everyone found it valuable.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

I'm excited to be here. Lovely conversation. Um, I mean, this is something I do and I'm exceptionally passionate about. I mean, I'm a builder at heart, so I'm excited for where we are. Um, you know, as an engineer, I am, you know, and a learner, I am, I'm super, super excited for what lies ahead and, and how much change it is, um, and, and I'm glad that this is a transformation that I am getting to see in my lifetime.

Conor Bronsdon:

Absolutely. And, and I definitely hear that passion coming through from you and hopefully our audiences as well. If you enjoyed this conversation with Arthie, I would say send this podcast to a friend who you think would also find value in it. This is a kind of a scary time as far as how fast this transformation is happening. Um, but it doesn't have to be. It can be exciting. It can be an opportunity. Uh, you can also get more info about Arthie, LinkedIn, and from our Dev Interrupted team at devinterrupted. substack. com and look for a follow up article talking more in depth about these topics coming soon. Thank you so much for listening everyone. And Arthie, thank you again for joining us.

aarathi-vidyasagar--she-her-_1_08-06-2024_140516:

Thank you.

People on this episode

Head Shot

Dan Lines

Host
Head Shot

Conor Bronsdon

Co-host