David Yakobovitch
You are listening to the HumAIn Podcast. HumAIn is your first look at the startups and industry titans that are leading and disrupting artificial intelligence, data science, future of work and developer education. I am your host, David Yakobovitch, and you are listening to HumAIn. If you like this episode, remember to subscribe and leave a review. Now onto the show.
David Yakobovitch
Welcome back listeners. Today’s episode of HumAIn Podcast includes Daniel Pianko¹, the managing partner at Achieve Partners². Daniel Pianko and I got to know each other through a few colleagues in New York City.
And in fact, as a principal data scientist at Galvanize, I found out that Daniel, you as well have been on our Galvanize campus in New York City when we had our massive location at the Hudson Square area and have been involved with Galvanize over the years. So thanks so much for joining us today on HumAIn.
Daniel Pianko
My pleasure. I think I was the longest serving person involved in Galvanize, actually, since 2013.
David Yakobovitch
It’s amazing to see the educational market. That’s gone through so many evolutions, as you mentioned, we look at a lot of these bootcamps, like the General Assembly and Galvanize. They’ve been around since 2012 and 2013. And now here we are, almost eight years later. We’ve seen so much consolidation, growth, development. I’d love to hear from your perspective, as someone who’s invested and built companies, what do you see about the educational market today?
Daniel Pianko
If there’s one thing that COVID is going to do, it is a massive experiment in taking millions of learners online in the space of a week. Almost every college has shut down its campus. Almost every K-12 school system has sent their kids home. Schools are just basically putting everything on Zoom, which is not the way to run a railroad. This is not the way to run online education. Usually online education, good online education, is heavily curated.
There’s a lot of different learning components that get brought in, both sort of pushed everything online. And I think what that’s going to do is actually interesting. Online education has sort of peaked in the US at about 30%. So roughly 30% of people were getting some of their education online and it basically peaked and it had a huge growth through the late nineties, two thousands. And then about four or five years ago, basically stopped growing.
You’re going to go from 30 to maybe 50% of people getting their content online. So you’re going to go through sort of a second massive evolution revolution in learning at all levels. As people realize that what they thought they could only do in person, they can do online. Even for us.
We have dozens of companies in our portfolio, in prior funds portfolios, and even ones that said, “Oh, we’re a hundred percent in-person, experiencial” in two weeks, they’re online. And that’s really innovation. But it’s V1 online. Everyone’s just put everything on Zoom. The ones that weren’t kind of prepared.
And as people realize they don’t have to be in person, they’re now going to make the investment, to make those online environments even more robust, even more like a replacement for the in-person. So the world can spend more of their time in more socially distanced ways.That being said, I don’t think in-person education is going to go away.
Universities are our societies’ oldest institutions. Besides the Catholic church, the first university was formed in 1088 in Bologna. And basically, ever since then, universities have been an integral part of the social fabric. I don’t think they’re going away. K-12 education and its modern format started about 150 years ago.
I don’t think it’s going to go away. If anything, the shift online is going to make these systems more robust and more prepared for the 21st century.
David Yakobovitch
As we look at the online technology, even at Galvanize, we’ve been making an extra focus for online programs. We’ve always had online programs, but now it’s been everywhere from enterprise to consumer, to government, focusing online.
And one of the really interesting things is I came across an article recently of an industry leader focused on online delivery. How do you deliver sales and education and all these different parts of a business. And what this leader said is that Microsoft Teams is not good for online yet. And I said, this is so fascinating. I dived into the article. And what it said is, when you look at software like Zoom and WebEx and go to a webinar, all of them have very powerful annotation tools, such as raising your hand, highlighting, dragging squares, laser pointers. Microsoft Teams doesn’t have that yet.
So it’s so interesting that online education is not just about the modality of being online and being either synchronous or asynchronous, but also what tools you are using to enable that learning.
Daniel Pianko
And the tools are critical. I’ve had this back and forth with a friend of mine, Michael Fertik, a Silicon Valley investor. And he’s basically like, why isn’t Ed Tech competing with Zoom? And the reality is Ed Tech isn’t going to compete with it. Apple is going to be the choice. Google, they’re going to be in the schools. It’s virtually impossible. Almost no Ed Tech platform has their own video interface for example, but Zoom is never going to build out the ecosystem that’s required to actually run an online school.
Let me give you one example from our portfolio: Packback. This basically grades online discussion boards. So almost every online education system that’s outside of Zoom, or better than Zoom, will require some interaction online between people, and Packback uses an AI system to basically put it up. Basically uses AI to allow professors to grade online discussion, because you’re not actually looking to grade very detailed work.
But you really want to make sure that students are actually engaging. They’re not copying from somewhere else. They’re not just saying yes as they post. The big criticism of online education was that in these online discussion boards a student just had to post, they just had to type “yes”. And then that would qualify as their post for the day. Well, that doesn’t quite work and it also doesn’t quite work to just copy something that someone else wrote.
So AI is actually really good and no professor wants to read through hundreds of comments. It’s not really a good use of the professor’s time. So companies like Packback are creating these tools that use AI to radically improve the experience, as the Microsoft Teams’ example. I deleted Microsoft Teams as soon as it got stuck on my desktop.
I don’t think any Ed Tech company is going to actually compete with Zoom or WebEx or any of these other well utilized systems. But instead, it’s all these tools around that platform.
David Yakobovitch
It’s so interesting, Daniel, that you brought that up about what Packback’s doing for the automation of grading for these discussion threads. Even at Galvanize, on our learn platform, we actually have something similar where we do some of that automation behind the scenes. We’re using things like containers. We’re using certain JavaScript-based scripts and we’re using certain packages in Python to do that with these ENVI greater systems.
And what I found so fascinating is that online grading and self-assessment and automation space really got started with turnitin.com many years ago. And a lot of us remember this from our high school and college days of essays and plagiarism. But now, as you mentioned with Packback leading the way, and a lot of other companies, especially as we continue shifting to online education. This is going to be ever more critical for teacher’s assistants and teachers, to be able to effectively go through material so they can support students with success, and not too many hours on things that could be automated.
Daniel Pianko
I remember we developed this for Galvanize. And you just go to Codeacademy and we invested a lot of money, Galvanize, in building this platform. And we were ahead of the curve and that’s why the Galvanize online learning environment was so successful and I hope it continues to be successful.
Most people don’t have Python coders on their teams. So Galvanize is unique because of its coding bootcamp. But think about your average K-12 school. Think about your average university. You’re not building enterprise grade software for those organizations. And so Galvanize is a unique example where we invest in their own. Most people are going to do it themselves.
I want to take one second. One of the most interesting and scary pieces to the move to online learning is that in general, online environments reward scale. Amazon.com, Google. The leader at a space gets two thirds of the market. The second gets 20% that’s being Google, and then everybody else shares like the last 20%. And you look at this across the US economy. You’re seeing technology bring massive consolidation. And that is happening in education because in order to make the investments at Galvanize, an online learning environment, you have to get scale, and scale is a different beast in the online world.
You look at universities like Western Governors University, Southern New Hampshire University. They are operating at a scale that was literally inconceivable 20 years ago in education. People criticize the University of Phoenix and others, and these big online universities, Liberty University, they are the future. And if you think about that and what’s going to happen in K-12, as people realize that, we’re going to have to move these things online and it’s gonna reward scale in a way that I don’t think people are ready for in the traditional education consumer market.
David Yakobovitch
Now speaking of K-12 education, and that means, everywhere from kindergarten through 12th grade in primary and secondary school for the listeners here on HumAin, some of you may be aware that as of late January 2020, Galvanize is now a K-12 company. So Galvanize was acquired by K-12 and a lot of what you’re saying, Daniel, is to help bridge the gap of in-person online education.
K-12 is a leader with operating charter schools, primarily online and over 35 states in the United States and supporting this, especially for different socioeconomic status students, especially those coming from non-traditional backgrounds. And the acquisition is so fascinating, especially because we’re looking today at COVID and if we bring it back there, we see even in New York City, for example, that public schools have gone remote through at least April 20th, which probably means the remainder of this calendar year and all exams and everything’s online.
And one of the big problems that we’ve seen that Mayor de Blasio and Governor Coumo have brought up is that the accessibility and equity for online education is not there yet.
And what they mean by that, first and foremost is, that there are at least a hundred thousand or more public school students who don’t have iPads or laptops to study school-remote because their parents are working remote and there’s no device or not everyone even has internet and wifi available at their homes.
So the partnership with K-12 is amazing because we’ve been focused on this online education and this movement towards “accessible for all” for a while. But I wanted to see your thoughts more so about that accessibility and equity that you’re seeing in the market, especially with online education.
Daniel Pianko
This is a huge issue. And I don’t think people quite realize how important schools, K-12 schools, physical schools, are in a lot, as I sit on the board of a K-12 school in the city. These places are where you get your food. Many of those hundred thousand, I forgot the numbers, but something like 50% of New York students are on free and reduced lunch, which means they do not have enough money for food.
And so they don’t have digital connectivity. And K-12 does a really good job, for example, and I give Nate Davis a lot of credit for doing the acquisition. It opens up a lot of opportunities for a lot of the students. He does a really good job in giving people the devices and the technology.
It’s a real issue, though, for traditional K-12 systems. New York should have been in a position to figure out online education for those hundred thousand students. This is very complicated, but it’s solvable. Whether it’s getting people devices, whether it’s opening up locations that students can go to, even in this time period. Many school systems have made the decision that because of, and this may be a little, hopefully, not too politically incorrect, but many school systems have made the decision. Well, if we can’t serve all of our students, we’re going to serve none of our students. And that’s a huge mistake.
Every public school system should be offering something like what K-12 offers to its student body 24/7, because, for many parents, they do need that content and it’s for everybody, to fall behind because you can’t figure out how to serve 10% of the student body is a massive mistake. And I would strongly encourage schools to look at organizations like K-12 and other online environments to figure out how to solve these equity issues. Especially, if it means getting technology in the hands of these kids.
It is not the right answer that, “Oh, these kids don’t have technology. Therefore, we can’t serve them”. The right answer is “let’s figure out how to get technology in the hands of these kids”. Maybe it doesn’t work for first graders or second graders, but any sixth, seventh, eighth grader, most sixth, seventh, and eighth graders, if you gave them an iPad right now or a cheaper Android device, they would.
We run a university network in Africa. It’s called Unicaf, with about 20,000, 30,000 students. Now, I know that there’s a founder of a company called Bridges. Bridges educates about three or 400,000 students in Africa leveraging very cheap Android devices. If they can figure that out in that environment, we can figure that out in the wealthiest, one of the wealthiest countries in the world, and one of the wealthiest cities in the world.
Seattle, the school district in Seattle made a similar decision. And I just think it’s the wrong decision. It’s a failure of leadership that we can’t get these devices and the internet connectivity in the hands of our students, and I know it’s hard, but that’s no excuse. And so the K-12 system basically gives every student in their program an iPad, is my understanding.
David Yakobovitch
That’s right. K-12 does offer the iPads. And in addition, for our listeners who are live or listening to us, some playback. New York state with the department of education did mention that initially was going to be a 25,000 to 50,000 iPad rollout. And now they are planning for at least 300,000 iPads to students in the New York state system.
Daniel Pianko
And that’s great. That is awesome. I didn’t realize they had done that. That is awesome.
David Yakobovitch
It is awesome, but it’s a little slow. So it’s like, we want to speed up the process and the key is not just the iPads, but it’s also the internet.
Recently, Google made an announcement that they’re changing YouTube globally to a default playback of only 480-pixel, not the 10 AVP that most people are used to. And part of this reason is a lot of YouTube live and different software even, including Zoom, has been buffering.
And some of that buffering has even been a two to three second delay. Can you imagine being in an online course and you cannot keep up with the professor because by the time you ask your question, they are already on to the next subject?
Daniel Pianko
And this is the failure. This is why it’s important that we differentiate between the aversion of online education that people are experiencing this week versus a real online education, because online education shouldn’t have to be synchronous.
A lot of your best learning, the best way to do online learning is to actually start with a very meaty question that a student has to wrestle with through a series of exercises, some of which can be synchronous, but many of which should be asynchronous. And so, instead of actually sort of being able to bring those online at scale, we’re bringing online a hundred percent synchronous, basically mimicking the in-person classroom.
And that’s where you’re going to see the buffering, the internet conductivity. The way you get around that is by parsing out and reducing the synchronous requirements and improving the asynchronous requirements. And so this is where the equity issues and the technology issues start to intersect, where you can actually create those environments much more effectively in a way that does increase access and equity.
David Yakobovitch
And part of that, as you mentioned, Daniel, improving asynchronous requirements allows learning to be more adaptable. It’s not just that we’re live all the time. Live online’s a fantastic model, but a hybrid model works just as well.
So whatever that percentage is, maybe it’s 20% or 40% of classroom-led instructor, live time with certain office hours, student presentations and support, but with an additional 50 to 60% of studying at your own cadence or pace, which in fact, can actually be a big win for a lot of students who, often, when they’re in the classroom, fall behind because we’re all different learning styles and preferences. But now you’ll have your time, your pace to keep up.
Daniel Pianko
And by the way, this can be done at all levels of education. When you come to our medical school, you get an iPad. And on that iPad is every lecture you’ll ever watch. Every book you’ll ever read. And then every time you come into class, you actually get asked a series of questions based on the reading from the day or the videos you watched the day before, it’s called Tiber Health.
And this can apply to early stage education, but also even up through doctors and medical school. And so, this and the way Tiber works, it’s really amazing. You come into class and this is actually a really good access story. We are kept at our medical schools with the number of seats that we can let people in.
We can only let in say 90 a hundred or about to move up to 150 people for a year. And we’re getting thousands of applications every year. And what we did was, we said, okay, for the people we get in the next hundred, 200 people, we’re going to let them in to the exact same curriculum because we digitize the whole curriculum.
And people would come in, sit in the same classrooms as the MD students or a slightly slower pace classroom. And then they take these exams within, because they were taking the exams on our iPad. We were tracking everyone’s results within three to six months with an R-squared of about 0.8. We can predict what your board scores are going to be.
So we were able to admit students who wouldn’t otherwise have gotten into medical school through this methodology, with this predictive analytics. It’s not quite AI. I shouldn’t use the term, even though this is an AI podcast, but using predictive analytics and sort of rethinking the entire educational process of training to be a doctor. We’re able to dramatically open the funnel, and if you think about how you get into medical school, you have to have gotten an A in organic chemistry.
The medical school we operate in Puerto Rico is effectively almost 100%. First-generation, low-income, $30,000 a year, bilingual. We graduated about 12% of the Spanish speaking doctors in the US, Hispanic doctors. And when we took over the school, before we implemented this curriculum, we had about a 65% pass rate on the boards. This year, we have about a 93% pass rate.
And what we did was we brought in this new curriculum, we ran people through this master’s program first that were on the fence that you have highlighted here. We could predict who would do well. And we were able to admit students from historically disadvantaged backgrounds into medical school, which is kind of the creme de la creme of higher education.
And what underpinned this was a total rethinking of the entire classroom experience, technology experience that led to a predictive analytics revolution in education, in medical school education. And now we admit students or we’re starting to admit students based on their success in the MSMS.
And the master’s degree is kind of a one-year program instead of the MCAT, because in the MCAT, needless to say, on many standardized exams, underrepresented minorities don’t do as well. And in fact, the percentage of underrepresented minorities in medical school is going down, not up at a time when our society is getting more diverse, and care in the language that you speak is incredibly important, and culturally competent care is so important.
So, you can transform equity issues through technology and through predictive analytics and through AI. And unfortunately, when, and this is, what’s really frustrating to me about how we’re entering this kind of wave two of online learning, is that it’s all on Zoom.
Zoom is the worst educational platform except all others. And so everyone’s adopting it, but it’s leaving out all these amazing innovations that should be driving quality, driving equity issues. And so I’ll get off my soapbox now. Sorry, David.
David Yakobovitch
It’s super relevant, again. Zoom is a good solution for webinars and for a lot of live learning, but not necessarily to be an end-all solution. And there’s so many platforms out there. We know from the canvases to other ones that are leading adaptive learning and the whole experience. A big takeaway from what you just shared is that online education is not only for software engineering and data science.
It can be across the board, whether it’s K-12, whether it’s legal, whether it’s medical, even business school, it can be online with the right tools, with the right processes and the right people set in place to help each and every student learn. And to Excel. And part of that is seeing what are the best learning profiles.
And that could be adaptive learning. I know this has been a big trend in the industry over the last few years, and there’s been some hits or misses in that space, but that’s come back up again, especially given the wave of COVID. Everyone’s saying, how do you make this the most adaptable to my learning style and my learning preference? What’s your take on the adaptive learning market and what you’re seeing right now with it?
Daniel Pianko
So adaptive learning has been the buzzword in education broadly, for the better part of 25 years. And even before then, some really great work was done down by very famous education professors who basically said there are different ways people learn, and that set off a generational pursuit, a multi-generational pursuit for creating content and curriculum that was more closely tied to, and geared towards the individualized learner. And we failed. So almost every effort to do this has failed. And I’m not a technologist, but what is important for tech hardcore, techies, to understand is learning is still one of those fundamentally human endeavors.
It’s actually the process of transmitting knowledge from one generation to the next. It is, frankly, what separates us from animals. And that is fundamental. If you think about how people learn, we have had correspondence courses, we’ve had libraries, you can get a college degree from a library for over 150 years in the United States. That’s not how people learn.
So there are huge gaping holes where literally, I was just telling, David, the other day, that the most well-funded AI driven learning environment with two, three, $400 million invested with the Silicon Valley royalty went bankrupt recently. And then another one got acquired for pennies on the dollar by McGraw-Hill.
So, we have failed. And the reason why is because the technologists and the educators aren’t connected enough. And I’ll give you an example of what has worked at Pearson’s math lab project, which has actually run, most successfully at Arizona State University. Actually, if you see where they deliver this, they actually deliver it in a big classroom with lots of computers lined up row by row.
And then you have a faculty member actually walking around, helping the students when they run into problems. And so you’ve got a totally personalized learning environment. But you still need the human intervention. And we’re AI driven, our personalized learning has been most successful in these entry-level classes.
But even in those environments, you still usually need some significant human intervention. And so, I think of the next generation. We’re not quite at the matrix where they just implant a chip in your head and suddenly you know how to fly an airplane. And we’re not where online education or AI driven education is totally worthless and meaningless. We’re at this kind of in-between stage where the most successful interventions are going to be those where the technologist and the education folks can come together and say, here are the areas where we can deliver a high quality program that radically improves the product and it’s going to be high-performance it’s like what I mentioned with that master’s degree in medicine.
It’s where you have all the tools available online. You have the predictive analytics underlying it, but then you actually come to class and you actually, and maybe in the COVID environment, move in fully online, but it’s still, basically, a Zoom class. So you do need that Zoom environment for a lot of stuff, but where you have both of these tools working together in a way that actually improves efficacy.
And that’s for the next investible period. So the next three, five, seven, eight, ten years, that’s going to be the focus. It’s like what Galvanize developed and what so many other successful organizations have developed. I’ll stop there, but that’s the future.
David Yakobovitch
That sounds right. And that is from the efficacy. What you’re sharing there, Daniel, it’s not just the live in the moment. Learning and helping students attune to their learning preferences, but when they’re not doing live. So when we’re looking at this offline learning or this asynchronous requirements, how do we customize and how do we make it set?
You’re taking skills assessments, and you’re constantly getting reminded on how you can improve or how the difficulties can change. I am a big fan of software engineering. And developer operations with DevOps, particularly with cloud platforms like AWS. A few months ago I took one of these cloud certified practitioner exams. And based on how you’re performing on the questions, the questions get easier or harder. And now even the SAT and the GMAT, and a lot of these tests do that, where they’re measuring to what level you can rise to that occasion.
Daniel Pianko
And those adaptive tests are a perfect example where technology works really well. Psychometricians can basically prove it to you. That’s a better model for testing because, as you can imagine, it levels out where you’re going to end up and allows you to drive a better outcome. And so you’re going to see what adaptive testing is, and that’s effectively what that type or example is. A lot of that is what the Galvanize stuff is. We’ll see that again, on the tool side.
I’ll give you another example on the tool side, which is quite on the straight education, but it’s where you’ll see AI have the most impact. We invested in a really cool company Google and Salesforce recently backed as well, called AdmitHub, which is a chat bot for admissions officers. Most questions that people answer in an admissions office are basic questions. How much does it cost? How many credits do I need? What happens on campus? And AdmitHub basically answers those questions in a chatbot format. And then it frees up the time for the admissions officers to help with more pastoral care, which is to help people coming.
There’s a big New York Times article that Bill Gates tweeted that featured AdmitHub because it reduced what’s called summer melt, which is people get into school, plan on going, but then don’t show up because life gets in the way, the funding dries up, mother gets sick, grandmother gets sick, brother has a problem. And so AdmitHub basically reduced summer melt by something like 30%. And what it did was, it freed up not the teachers, but the administrators who help students to think about higher level issues by having this chatbot answer basic questions.
And so, while the actual instructional component will stay fairly human centric for the foreseeable future, a lot of these back office, I don’t call the admissions office back office, but sort of these non-straight academic functionality will become much more consumer-friendly and tech-driven and where AI can have a massive impact.
And so, for your audience, my advice would be actually stay out of the classroom. Don’t try to change how people learn. People are learning the same way for a couple millennia. Everyone’s reading sapiens these days. And our brains are pretty hardwired by history, but where AI can really have an impact?
If you’re not an educator or you don’t have connectivity to the education side, it is these back of the house functions like admissions. We have a really interesting couple of EverTrue whose functions help development officers. So people fundraise to make them much more effective.
So we’re big believers in AI and education, just not necessarily in the classroom tomorrow, or if it is going to be in the classroom, it’s gotta be more than likely going to be human intermediated, more than likely going to be hybrid, at least for the foreseeable future
David Yakobovitch
And this hybrid learning is very effective. As we’ve been describing today on this episode of HumAIn, learning is possible with good outcomes and learning can be possible both online and in person. And what we’re starting to see is more of these technologies, like the AI chatbots, particularly from AdmitHub that can help students have better outcomes.
There’s so many educational companies in this space right now, especially even in ed tech. On a future episode of HumAIne, we’re going to be talking to Ash, who’s from the ed tech Accelerator in New York City, about ed tech week, which I go to every year. I’ve met a lot of founders and see what they’re doing with new tools, technologies, and processes. And I know that conference is moving online. As many conferences have been, as a result of COVID, but some of this can be good. Again, it can add more accessibility and equity to the space.
In fact, one of the leading pioneers in the software and data space with training and books and a lot of that material for self-paced is called O’Reilly. O’Reilly media have recently made the announcement that they have gone completely online with their conferences. And they’ve actually completely shuttered the in-person conferences business indefinitely, which I’ve found quite surprising. And they indefinitely took that position in the last week or two. In-person will still work. And they were going to come back stronger from where we are today, where we’re going to have both models because , quite honestly, and frankly, some people prefer in-person learning and there are those different preferences.
How about yourself, Daniel? Are you more of an in-person or online kind of guy?
Daniel Pianko
Oh, don’t ask me personal questions. I can’t handle them. People learn differently in different components. Sometimes I actually really prefer online learning. I mean, I actually do consume a decent amount, whether it’s figuring out. I just had to figure out something in the house.
And like here I was Googling how to fix it. Anyway. I won’t go into it. But whereas, I find conferences. I actually don’t attend many of the sessions. Usually I find conferences really great for interacting with people and being able to set up a large number of meetings in a row.
And so I’m actually not a believer that COVID is going to radically change human existence. The last hundred years we have not had a, I tried to explain this to my kids. We have not had a pandemic, a real true pandemic in about a hundred years since 1918 with the Spanish flu. Which by the way originated in Kansas City, but let’s not get into politics here.
And we’ve been very lucky as a society that we have not had a pandemic in about a hundred plus years. Pandemics were the norm up until recently. Pandemics were every five years or so. There’d be a major pandemic throughout Europe up until that 150, 200 years ago. Isaac Newton discovered gravity while quarantined during the bubonic plague outbreak from Cambridge. And by the way, during that bubonic plague outbreak, 25% of the citizens of London died. To put that in perspective, if that kind of a plague hit New York today, New York City alone, you’d have 2 million dead. So we have lived in a very unique time where we haven’t had to deal with these types of things.
But despite our history of plague and quarantine, we’ve always ended up coming back together, as people, in groups. And I don’t think technology fundamentally changes that. And so this may be too esoteric for this podcast. But I do believe that the vast majority of humans want human to human interaction.
And I do believe that science will figure out, even in an age of pandemics, there was still human to human interaction. And I believe that scientists will figure out how to manage whether it’s COVID or the next one. And hopefully the next one is another a hundred years away. And O’Reilly will restart conferences once there’s demand for it.
And there’s going to be demand for it much sooner than people think. If you look at those pictures coming out of Wu Han right now, a hundred percent everybody’s out and a hundred percent, everybody’s getting back to their relatives.
I was talking with the CEO of a company. Most of this is deferral. Not loss. If you’re going to get married during this time, you’re still going to get married. You may have gotten to get adjusted in the piece, but you’re still going to throw a party for all your friends. Every single conference, maybe they postpone it for this year. Maybe O’Riley is unique. Everybody’s going to want to be with their friends, be with their families. The number of post COVID raves that is going to be huge in 60, 90, 120 days. And I haven’t seen my parents in three weeks. They want to come visit. I’m telling them no. So, everybody’s going to want to come together.
And so, I’m long “human interaction”. Maybe not the right format on this podcast. And it might change. It might evolve, but humans want to be with other humans. So I’m definitely an in-person learner. But I’m also an online learner.
My kid, and this isn’t a generate. Fine. I’m 40, whatever. My kid who’s 10 hates online. All he wants to do is go play soccer with his friends and all the rest of the kids. He doesn’t want to learn online. And my daughter doesn’t want to learn online. They want to be with their friends. Again, I defer to you and the other technologists on this podcast, but I’ll take the over-under that in 10 years we will have as much, if not more human interaction than we have in the past.
David Yakobovitch
I’m with you on that long bet. Sure. Some of it may be digital, but we’ll definitely be moving to human interaction even more so, but there is something else huge, and that is the huge skills gap, which will be our final topic for today’s show.
So we’ve seen a lot in software engineering and data science and different skills that need to be bridged over the last 8, 10 years that’s become a talent war, especially between major cities and Silicon Valley. I recently published an article towards data science that states the fastest growing job in 2020 is cybersecurity analyst. This is one of the most underappreciated fields around privacy and security. And we’re even seeing that now where there’s different state actors attempting to perform hacks and we’re seeing more deep fakes coming online with the election 2020, just around the corner. Do you also see cybersecurity as a fast growing field or where’s your take on where the biggest need with the skills gap is today?
Daniel Pianko
Well, first of all, the skills gap is massive and it is not going away. Even with the projected unemployment from COVID. There are 7 million unfilled jobs in the United States today, and that number is really driven by a couple of things. One, we have two types of frictions. We have education friction. Which is that people don’t want to get educated because it costs a lot of money. It takes a lot of time. And most importantly, you’re not guaranteed a job.
So even if you learn cybersecurity and you’ve got your SOC-1 certificate, you’re still not guaranteed a job. On the flip side, we have employers. And employers also have frictions. They don’t want to hire anybody unless they’ve done that job before. They want people who have all the skills to be successful and hit the ground running.
And so between those two, you’ve got sort of this massive chasm between education and employment. And I want to call out technologists on this one. Google applicant tracking system, AI. What has changed in the last 10 years is that because of the onslaught of resumes hiring managers have turned to applicant tracking systems to sort through the wheat from the chief.
And unfortunately, what this means is you can only say “good people skills” so many times in a job record. So instead, what everyone’s doing is just adding more and more skills. Bachelor’s degrees, which may or may not be relevant or SOC-1 analyst, which may or may not be relevant or SOC-2. You’ve probably had your certification, all these different certifications that you would need.
And the reality is most of those things can be taught relatively quickly, but we, as a society, have abandoned teaching those skills. And we have left the teaching of these, we call it last mile training, oriented skills. There aren’t really pathways to employment in these areas, whether they’re cybersecurity or certain areas of health tech or linemen, people who string the lines up to make sure that your phones and internet work.
So you’ve got huge, vast areas where the connectivity between education and employment has broken down. And we see a future where a series of intermediaries develop in cybersecurity, in software development, in electronic medical records, in programming pacemakers once they’re in a beating heart, in all these areas where there are very specific skills. Salesforce management, there are 300,000 unfilled jobs for Salesforce managers in the United States.
And the secular shift to software eating the world is driving a lot of these trends. And I don’t think they’re going to go away despite COVID and everything else. And what we need to develop are a series of, we call intermediaries, that solve the education friction and the employment friction.
These are organizations that will employ large numbers of people, and will train them at the same time. These are folks that will say, take the Galvanize example, we’ll let you into the bootcamp, but then we’re going to guarantee you a job at the end, because we know that UPS is going to hire 30 Python developers with this skillset.
And this is a lot of what we’re doing right now. Our fund is basically a hundred percent focused on buying potential intermediaries and adding last mile training. And we’re looking for these big scale gap areas. And coming back to what the largest one is, sure, cybersecurity, is going to be huge. No doubt that especially with everybody working from home, there’ll be more and more need for cybersecurity analysts. But what is the pathway if you’re sitting at home and you’re a recent college grad to getting one of those jobs? And right now the answer to that is, well maybe take an online course.
Maybe there’s a company called SecureSet, which has been doing a lot of this recently, just got bought by Flatter, and there’s a couple of bootcamps, but we need hundreds of thousands of these people. And they need to be American citizens to work for governmental agencies and there are all these others. So why aren’t we recruiting recent college grads into large scale training programs where they’re working while they’re learning?
And that that’s going to be the future. And the skills gap is going to be with us for a long time. It’s more than just cybersecurity. It’s not just cybersecurity. We’ve identified about 50 different areas where this kind of supply demand imbalance is broken with education.
David Yakobovitch
It’s incredible to think that there are over 50 different areas for this imbalance, with education. Traditionally, we think of software engineering, data science, full stack development, cybersecurity, cloud, and web ops and DevOps.
Daniel Pianko
Well, you just mentioned like six.
David Yakobovitch
But that’s not 50. These are like six.
Daniel Pianko
Those are the six you know about because you’re in those fields. Now think about a hospital administrator. If you want to think about what’s happening in hospitals right now, think about that like a cath lab, do you know what a cath lab is? It’s where if you have a heart attack, they come in and they stent your heart. And talking to the largest staffing company for staffing cath lab techs, it’s not just the nurse. You’ve got three or four levels of people who are involved in stenting someone’s heart. So he’s got 700 unfilled jobs and he’s basically saying, look, every time there isn’t one of these positions at a cath lab, the cath lab has to close down.
So that means if you show up at a hospital and they don’t have a cath lab tech, certified to X level, they’ve got to shut the cath lab and people die when that happens. Because you have a heart attack, you can have a heart attack at midnight and you’d go to the hospital. And they said, sorry, we don’t have the guy who stents your heart that way, that would be a bummer for you.
And so, healthcare and technology are massive areas. But, if you look at areas like energy and power, and these are not all white collar cool, cushy jobs. HVAC, people who repair your home HVAC or small commercial HVAC system, they have pretty sophisticated skills.They get paid 80 or 100 thousand dollars a year, but we have a massive shortage of them.
There are 50, 60, probably a hundred plus areas where this pathway from education to appointment or even just, Hey, I want to get, I need a job to feed my family. There are a lot of these very significant areas where you’re going to see intermediaries develop. It’s not just in tech.
David Yakobovitch
So from these 50 areas, from the trends, from everything that we’ve been hearing about on today’s episode of HumAIn with you, Daniel, what is your call to action or what’s next that some of our listeners should think about?
Daniel Pianko
That’s an unfair question because there’s so many calls to action here, but, the most important thing for this audience is software is eating the world and it’s changing how everybody operates. But at the end of the day, things around education and workforce are very human-driven. And there’s a push to say, Oh, we’re going to automate the job search process, or there’s a push to automate the education process.
But at the end of the day, most people hire people they know. And even Amazon, which has to hire a hundred thousand people, they are still primarily network-driven organizations and human network-driven organizations. So I would encourage you as you think about the software you develop. Everyone wants to get into education or workforce to realize that these are some of the biggest decisions people make in their lives. And they still make them based on very human interactions with other humans. And that’s beautiful on one hand, but complex on the other.
David Yakobovitch
Daniel Pianko from Achieve Partners. Thank you for being with us today on HumAIn.
Daniel Pianko
My pleasure. Thanks, David.
David Yakobovitch
Thank you for listening to this episode of the HumAIn podcast. What do you think? Did the show measure up to your thoughts on artificial intelligence, data science, future of work and developer education? Listeners, I want to hear from you so that I can offer you the most relevant transcending and educational content on the market.
You can reach me directly by email at david@humainpodcast.com. Remember to share this episode with a friend, subscribe and leave a review on your preferred podcasting app and tune into more episodes of HumAIn.
Works Cited
Companies Cited