Conversations on Applied AI

Mona Akmal - Using AI to Predict Efficient Revenue Growth

Justin Grammens Season 3 Episode 7

The conversation this week is with Mona Akmal. Mona is the CEO and co-founder at Falcon AI. Falcon is the first intelligence platform for the entire GTM team, which includes marketing, sales, and account management. She is a product and engineering veteran who builds resourceful, kind, and output-driven teams at scale, with a passion for crafting elegant solutions to technically difficult problems. She has grown teams, businesses, and many products at Microsoft, Code.org, and Zulily. 

If you are interested in learning about how AI is being applied across multiple industries, be sure to join us at a future AppliedAI Monthly meetup and help support us so we can make future Emerging Technologies North non-profit events!

Resources and Topics Mentioned in this Episode

Enjoy!

Your host,
Justin Grammens

Justin Grammens  0:00  

Greetings Applied AI Podcast listeners. This is Justin Grammens, your host of the Conversations on Applied AI Podcast. Just dropping in to let you know about a very special event we have coming up on Friday, May 12. It's the spring 2023 applied AI conference. You can learn more by going to applied AI conf.com. This full day in person conference is the only and largest artificial intelligence conference held in the upper Midwest. It will be in Minneapolis, Minnesota on May 12, we will have more than 20 speakers with two tracks covering everything from Ai, business applications, ChatGPT, computer vision, machine learning and so much more for being a listener to this podcast, use the promo code podcast when purchasing your ticket for a 50% discount. So here's the details. Go to Applied AI Conf dot com. That's applied Applied AI Conf dot com, to see the full schedule and register for the only and largest artificial intelligence conference in the Upper Midwest on May 12. And don't forget to use a promo code of podcast when checking out to receive a 50% discount. We look forward to seeing you there. And thank you so much for listening. And now on with this episode.


Mona Akmal  1:08  

When I talked to companies, and they told me that they have seven data scientists who build their revenue forecasting models, my first question is, are you in the revenue forecasting business? And they said, no, no, no, we do influencer marketing. Okay, great. So then why do you have people building revenue for gas models for you? Why not apply it as very scarce and branches resource to deepen your own competitive modes and make your Cournot offering compelling? So my ask is that we not just apply AI to business problems, but we apply them to business problems that are uniquely ours to solve and actually create a deep and differentiating mold for our business and for that, the AI community within corporations needs to understand what business they're actually in.


AI Announcer  2:00  

Welcome to the conversations on the applied AI podcast where Justin Grammens and the team at emerging technologies know of talk with experts in the fields of artificial intelligence and deep learning. In each episode, we cut through the hype and dive into how these technologies are being applied to real-world problems today. We hope that you find this episode educational and applicable to your industry and connect with us to learn more about our organization at applied ai.mn. Enjoy.


Justin Grammens  2:31  

Welcome everyone to the conversations on the applied AI Podcast. Today we're speaking with Mona Kamal. Mona is the CEO and co-founder at Falcon AI. Falcon is the first intelligence platform for the entire GTM team, which includes marketing, sales and account management. She is a product and engineering veteran who builds resourceful kind and output driven teams at scale, with a passion for crafting elegant solutions to technically difficult problems. She has grown teams, businesses and many products at Microsoft, Code.org and Zulily. As VP of Product engineering at Code.org, she was responsible for more than 100 hours of computer science curriculum being accessed by 100,000,000 K-12 students in the US and globally interesting stat is that more than 5 billion lines of code have been written by students on a code.org website since its inception. So that's that's an amazing accomplishment. Mona, thank you much for being on the program today.


Mona Akmal  3:18  

It's great to be here, Justin. Awesome. Well, good. So


Justin Grammens  3:21  

I talked a little bit about, you know, sort of some of these organizations that you've worked with along the way, you know, maybe you could rewind back and say, you know, sort of think back to how you got into this this field. And maybe what was the trajectory of your career?


Unknown Speaker  3:33  

Yeah, absolutely. I actually like to start with a story I used to share with people like.org Because there's such a strong reservation kids had about getting into computer science. When I was in school, I skipped all computer science classes, because I thought it was something only nerds did. I love math, but I didn't really care much for computer science. And I was actually much more focused on AR, but then my dad died. And you know, that was a big reality check. I was 12 years old, and we had no money. And so I had to grow up really fast and make a pragmatic decision about picking a line or discipline where I could be financially independent, take care of my mom. And so I wish I would tell you I loved computers since I was a kid. For me, this was a decision about gaining financial and social independence. And so I got into computer science. The first time I engaged with computers was when I was an undergrad. The reason I share this story is because people are terrified of computer science, especially women. And I know if I can do it, I really believe anyone else can as well. So I did my undergrad in computer science and then got hired as an engineer at Microsoft. And that's essentially how my story with technology began.


Justin Grammens  4:56  

Excellent. So you got hired at Microsoft, you were doing software developer And is that true? 


Mona Akmal  5:01  

I was I was an engineer in the GDI team, which is the core graphics layer of Windows. So you know, when on your screen, you see a line being drawn, or circle or a font that is all written by GDI. So it was a very heavy maths plus C++ programming type of job.


Justin Grammens  5:21  

Gotcha. And maybe as your career sort of evolved, and you started working at a couple of different other companies and got involved, and of course, I want to talk a fair amount about the code.org experience, too, because that's just, I mean, I, I teach a university course, but we also run this applied AI Meetup group, you know, no one's getting paid for any of it. We're all just very interested in technology. And so I really, really love that side of your career. But you also somehow kind of got away from maybe the fingers on the keyboard and the programming stuff more into a product role. Is that true?


Mona Akmal  5:50  

Yeah, absolutely. So you know, this is something that's been very true for me, I like being great at things. And I was an engineer for three years and realize that I could be a good engineer, I couldn't be a great engineer. And, you know, no amount of work or effort would get me to that level of greatness. However, I am a natural product person, because I tend to be very curious about how people do things, and finding better ways to do the same things and saving time so that people can do more with their lives. And that's when I moved into product management, and then got into general management. My last role at Microsoft was running the product management team for one write applications. And then that's Microsoft to help three very different CEOs build three very different companies. And that was for me, essentially, CEO school,


Justin Grammens  6:40  

you know, you kind of been getting more and more into data, it looks like you know, as I kind of look over your your career, like product usage data. And I think we'll move more into your into your startup that you're working with as well. But yeah, I mean, it seems seems like that was an important piece of whatever you were working on.


Mona Akmal  6:54  

Absolutely. I'm a big believer that the environment that we all find ourselves in being relevant is really important. And I'm always looking for how can I upskill and retool myself and my own knowledge about the space so that I remain relevant, because to be relevant is to be impactful, right? Things that I learned in undergrad 20 years ago, if I was still stuck there, I would have no impact in the world at all. And so the choice to focus on data and machine learning applications is simply a pragmatic one again, which is, to me, it's one of the key big evergreen themes within technology, where innovation is going to continue to happen rapidly. And being at the forefront of that is is super, super important.


Justin Grammens  7:43  

Excellent. How long has Falcon been around?


Mona Akmal  7:46  

We've been around for three years, three years.


Justin Grammens  7:48  

Okay. Maybe you can elaborate a little bit more. I mean, I mentioned it during the beginning. But you saw a need in the market in this in this GTM space? What were some of the things that you saw, and that maybe led you to want to launch this this new company?


Mona Akmal  8:01  

Yeah, so actually, the inspiration for Vulcan came from my time, that's really it's the only place I've ever worked, where I've seen data actually used for operational intelligence. And that's because it's an ecommerce company, right? We were not selling software where margins are 100%, we were selling leggings for little kids where margins are very, very razor thin. So you have to run a very lean, highly operationally rigorous company. And so on the one hand, I observed that our customers were young moms would get these highly personalized emails and experiences that drive engagement drive them to purchase products. On the other hand, none of those personalizations were being handcrafted, they were all automated because we had learned how to use our data to really deliver an E plus personalized customer experience to our customers. Then when I moved into b2b SaaS, I was honestly shocked to see how little data is used even by the fanciest of tech companies. And we're still stuck in the dark ages, in my opinion consumer is lightyears ahead of b2b SaaS in terms of using data for day to day work. And so the opportunity I saw in the market was bringing some of the intelligence that I gathered from the consumer world and how data is used to drive customer experience and revenue and apply that to the b2b SaaS world. And what was specifically interesting there was also, as we've all seen, even b2b SaaS companies are now moving into models that look more and more like how consumers buy software, right? A lot of business applications are now sold for free. So for instance, if you wanted to start using slack for your work environment, you don't have a conversation with a sales rep at Slack. You just download and start using it and then down Have a line someone from slack will probably contact you and say, hey, we'd love to hop on a call with you and tell you about how or why you should convert and become a paid customer. And so that was really the inspiration behind Falcon was helping b2b SaaS companies come into the 21st century in terms of using all their data to drive revenue,


Justin Grammens  10:20  

Hasnain that's good. And kind of running a b2b company here on my side, sort of in my day job. I know exactly what Jimmy I mean, there's a lot of ways that I joke with my marketing crew here because like, I feel like the local gas station down the street, well, it's an auto repair shop, they send me all sorts of information that is actually highly relevant to my car, they send me coupons. I mean, these are things that are kind of like, but kind of marketing one on one it feels like but I'm like, Why aren't other businesses doing this to each other? You know, a lot of ways. And these are things that I sort of realized as a customer, I'm like, Oh, that was a nice little touch. You know, I'm glad that you followed up with a survey for me like does this is don't do enough of that stuff.


Mona Akmal  10:58  

That's correct. As consumers, we demand more, because we have been trained by our own experiences interacting with consumer brands, right? So on the one hand, we're getting these highly personalized experiences, like you are from the auto shop. But then on the other hand, we're being constantly disappointed in the work setup, when we are acting as business buyers, or business users of software. That's the problem that needs to fix.


Justin Grammens  11:23  

Yeah. So how are you then applying AI and machine learning to to some of these problems that we're seeing here,


Mona Akmal  11:30  

this might be a contrarian view, a lot of people have this sense of mystery and magic around AI. As you know, most machine learning applications are actually a data churning problem and a data cleaning problem as opposed to a machine learning problem, right? So Data Prep is the biggest part of the machine learning pipeline, that is the blocker. So in a nutshell, what we do is we join sales activity, data marketing, activity, data and product usage, activity data in ways that are deterministic, and heuristic. So we use some of machine learning in order to do that. And then we try to determine key insights that drive revenue. For instance, if you notice that a particular account, their usage of your product is anomalously, or consistently trending up, or trending down, could you in for that insight by looking at that time series data, again, using different machine learning techniques, and surface that to the person that owns that account so that they can go and have a timely conversation about expansion or provide minimizing churn risk. That's just an example of something you can do in South Korea. And it's an example of how we think about applying machine learning and AI. Another example would be not just the ability to do forecasting for your revenue, but also to be able to do forecasting for expansion. And being able to predict churn and being able to even predict how many meetings you're gonna book in a quarter, because those are all leading indicators for how much revenue you're going to generate a 12 hour exam, right?


Justin Grammens  13:12  

Sure. Sure. Yeah, you guys kind of, like you said, that data cleaning problem, do you guys handle a lot of the cleaning aspect of it? We do. And so I would take my marketing system, I guess, my sales system, my CRM, you know, all these sort of pieces together and sort of integrate it into your into your platform?


Mona Akmal  13:29  

That's exactly right. Okay, we would pull that data in. And then we would use some machine learning techniques, as well as some classic sort of relational techniques to join that data. So that is one application of machine learning. But there are several others as well, that are downstream.


Justin Grammens  13:45  

Gotcha. Okay. Yeah. And we'll have links to your website, and, and all that sort of information here in the show liner notes. I see there's a way to get a demo on your website. I mean, you guys seem to be a very, very visual company. I mean, I like the colors are vibrant colors. But you vote Yeah, I mean, at the end of the day, I think you want to provide a lot of I would assume transparency to sort of like what the system is doing and where you should sort of spend your spend your dollars.


Mona Akmal  14:07  

Absolutely. And you know, our color choices, I think are also exemplary of what we're trying to do, which is traditionally b2b brands have been boring blues and boring greys and boring blacks and purples, or they go super kitschy, right, like with gone being magenta and hot pink and whatnot. And we're trying to bring some of brand sensibilities standpoint, a fun, but mature point of view, which is why we're not afraid of color, just like consumer brands aren't afraid of color. But we're also not like kindergarteners, who, you know, gotten their hands on the peons for the first time, either. Good. Yeah. So how big is the organization? We are now 32 people. Wow.


Justin Grammens  14:53  

And are they all kind of a virtual based company sort of scattered all over the world?


Mona Akmal  14:58  

We are headquartered We're in Seattle, and our product and engineering and science team, which I, obviously I'm biased. We're a product led company is based in Seattle. But we have people all over the US, our sales team, our marketing team spread all.


Justin Grammens  15:14  

Sure. I mean, one of the things that I think I saw on your bio was I mean, you self funded this, is that is that right?


Mona Akmal  15:21  

I bootstrapped for the first year. But that time, I did not want to bring anyone else on. So I spent a year by myself, and planned for that financially and Bootstrap, and then my co founder join. And we spent about six months building things out ourselves. And then we raised a seed round from Greylock and pillaging. I


Justin Grammens  15:45  

see. And so how's it been being a woman tech founder, some pros and cons, things you've run into?


Mona Akmal  15:51  

You know, I personally don't relate with myself as a woman tech founders. So it's always interesting for me to process that. I have noticed though, that in general, when I talk to my friends, who are also founders, it's not so much that we are assessed differently, it's that the questions I am asked versus the question they're asked are different. So I am asked for more evidence and more proves that what I'm saying is fruitful. So I'm asked for more traction than my male counterparts would be they're asked more about their vision. There's a statistical research on this as well, that men are asked about their big bold vision and women are asked about proof and evidence that they can actually do the job. However, you know, I've been in the tech industry for over two decades. So I feel like I'm already pretty trained and accustomed to be 10 times better than the average male founder in order to get the same outcome. And one of the reasons why I started Falcon was because in order for us to change the conversation, we need more women making personal sacrifices and not yielding to the systemic bias that shows up in their lives every day. Is it fair? No. Do I give a fuck? Not really. Because it's about winning. It's not about fairness, right? For sure.


Justin Grammens  17:18  

Interesting perspective, for sure. Ya know, so you're, you're just sort of out to out to make things happen. I mean, I think was one of the things you maybe said, you know, earlier around data, I guess, and and making sure that you want to make a change, you want to be impactful? I think that might have been the word that earlier.


Mona Akmal  17:33  

That's exactly right. I think we all need purpose in life. And for me, this is a purpose I can't be repaired.


Justin Grammens  17:40  

Well, that's, that's excellent. What Where do you see the company now going, then in the next, say, three to five years or so? Yeah, so


Mona Akmal  17:47  

we just raised a $16 million, Series A. So now much like any venture backed company, it's all about taking the thing that we build with a very founder led sales, motion, Mona knocking on doors, getting customers to learning how to build our own go to market function, and being able to work through our sales and marketing team instead of me. And obviously, you know, because the team is scaling, learning how to operate in this new environment and getting to being a profitable business with some unit economics, that is really the focus of the team for the next 12 to 18 months. And then when we feel that our go to market engine is ready to, you know, set some fuel on, then we'll go get more fuel, we'll get more dry powder and scale the business past that.


Justin Grammens  18:41  

Sure. So how can the applied AI community help you in that? In that way?


Mona Akmal  18:45  

I would say a few things one, build versus buy is an interesting conversation, I see a lot of data scientists within companies wanting to optimize for the projects that they're excited about, as opposed to enabling the business in the fastest way possible so that the business succeeds, right? My my pet peeve with the applied AI community is not everything needs to be built in house, when I talk to companies and they tell me that they have seven data scientists who build their revenue forecasting models. My first question is, are you in the revenue forecasting business? And they said, no, no, no, we do influencer marketing. Okay, great. So then why do you have people building revenue for gas models for you? Right? Why not apply it as very scarce and precious resource to deepen your own competitive moats and make your Cournot offering compelling? So my ask is that we not just apply AI to business problems, but we apply them to business problems that are uniquely ours to solve and actually create a deep and differentiating both for our business and for that the AI community within corporations needs to understand what business they're actually in. So stop needing to build everything from the ground up, you know, be pragmatic about where you should buy solutions versus where you should build.


Justin Grammens  20:15  

Yeah, that's a great perspective. I see you guys are hiring on your site. So definitely can drive some people there. I mean, it feels like it's, you know, your marketing manager and sales and stuff like that, or where do you see AI, like AI is being applied within your business in the next three to five years, we


Mona Akmal  20:31  

have a really sent knowledgeable science team, and we have research scientist, and then we have a few applied scientists and we work together, our research scientist is really looking at bets that are going to deepen our strategic moats, six 912 months out, and then our blind science team is really focusing on bringing the light for existing customers use cases, features and capabilities that help them. So churn models, expansion, cross sell upsell models, forecasting models, time series analysis, we do a lot of that work. I would say, as I mentioned, for me, I am not going to hire a scientist for our internal data, because we use Falcon for Falcon, so we don't need it. Right? We will only because science is so expensive, it is high risk, and it's high reward potentially, I would spend every science dollar in our budget towards deepening our products and our product differentiation. And so that's really where we're going to be spending our science dollars. Yeah, good.


Justin Grammens  21:41  

Sounds good. I mean, are there things coming up in the technology that you maybe see on the horizon that you think you could directly apply? Or is it you feel pretty good about the state of it today? And there's a lot of stuff that you can just do with with sort of what's already being done with artificial intelligence?


Mona Akmal  21:55  

Yeah. So I mean, I think that a lot of time series analysis, there is fantastic research that has been done. And we've applied quite a bit of it in power. And as well, a few of the problems that I don't see any good solutions for in the research part, and we don't have the time and bandwidth as a young company is being able to handle seasonality at a multi dimensional level. So you know, all the packages that get there, okay, and detecting seasonality at the highest level. But the more you got leaner, the less accurate it gets. So I think that's an area where I would love for more research to happen. And for it to be accurate research, we actually chased a few different research papers and realized that they were just garbage papers being written for for the sake of credit, but not really, there wasn't any real science there. Another thing that I'm excited about is the application of ChatGPT. Within the sales community specifically, the chat bots that are on websites are garbage between you and I, they're not very cool, really thinking about how can we automate some of the outreach that sales reps do that sales development reps for prospects by using something like ChatGPT instead of relying on human beings? Again, this is all about driving efficiency, by I think the unit economics on that have to get better, it's still pretty expensive. We have to sort of see how you apply that over the next 12 month.


Justin Grammens  23:28  

Well, I mentioned code.org. I'm curious to know a little bit about how you got involved with that organization.


Mona Akmal  23:33  

I wish I could say it was a very thoughtful thing. I sort of fallen into it. I left Microsoft knowing that I did not want to work in a huge company because my strengths and superpowers don't really shine in 100,000 person company. And I was looking very much to go small. I proved to myself that I still had it that I could still build meaningful products without needing a whole machinery around me. And I wanted to retool because, you know, back when I left Microsoft, Microsoft was not considered a great place to work. Right. This was like Steve Ballmer had just left. Satya hadn't even been announced as the CEO back then. And so I knew my skills were outdated. You know, when someone would say slack, I would say what, like, I didn't know slack. I didn't know JIRA. I didn't know Trello. I wasn't windows. We never used agile, we had a three year product release cycle, right? And so that's when I met the founder of code.org. And he had just recently done this video that had gotten 40 million views and Hadi Partovi and, you know, she was looking to build that into a curriculum, use that demand and that amazing responsive documents are needed do a curriculum for kids, and I met hottie I mean, I thought I was meeting a verb Aussie, he asked me to reverse a linked list, so I did. And then he said, Great, while you're figuring out what you want to do with your life, why don't you come in, you know, run product and engineering here. I said, Sure, because this is a topic I do care about, I actually had my own nonprofit in the education space a while back. And so that's how I got into it. And it was a phenomenal experience. Because one, as a product person, designing software for little kids is possibly the hardest thing you could, they are very difficult to please, their attention spans are very short, if they don't understand something, they're not going to try to a whole lot longer, they're just going to give up. And so you know, we had the most hilarious user research studies where I don't have kids. So I don't really know how to deal with children. I had a bunch of eight year olds, and I accidentally gave them all juice boxes. And so there was a sugar high. And then there was a sugar crash, and I had kids running around screaming. So I learned a ton about children, about building software for little kids about how nonprofits work about all the biases that gives have about why computer science isn't meant for them? Yeah, so it was just a beautiful experience.


Justin Grammens  26:16  

That's amazing. From my recollection, you know, and Co dot Oregon, it's, you know, here's a whole bunch of resources that anybody can come and use. And there were certain, you know, sort of days throughout the year on the calendar, where everyone was sort of trying to come together, and let's all code stuff at the same time. But it seems to have evolved much more into that. And, you know, I guess I didn't even know there was a title of VP of product and engineering, you know, there. So you had a chance to really sort of put your mark on it. How long were you there? And sort of like, what is your organization doing today? I guess, to your knowledge,


Mona Akmal  26:45  

yeah, actually, I mean, I sleep pretty close to like, woot.org, transact, actually. And so I was there for a year and a half. And when I joined, we had a video, we didn't have a curriculum, when I left, we had built a K through eight curriculum. And we were building the eight through 12 curriculum and working with, you know, some of eight of the largest school districts in the US like Chicago Public School System, the LA public school system, and so on. Now, as of today, not only is called.org deeply entrenched in our school systems in the US, but it has also expanded globally where the curricula have been translated and modified to work offline, to support computer science education, again, eight through 12 was the primary focus in the developing world and South America and so on. And, as always, so code.org was not just it was a curriculum company, we built curriculum, interactive curriculum. It was a training and professional development company. So we trained teachers and teachers of teachers, because with computer science, one of the biggest problems is there aren't enough teachers who know computer science well enough to teach it. So there's a chicken and egg, right. And then the third thing code.org does is a lot of advocacy work, which is lobbying in Washington, DC, to make sure that computer science credits called towards science and math in high school, right. And so a lot of advocacy work. And then there's a lot of bottoms up awareness work, which is what the Hour of Code is all about, is to dispel all the myths about why computer science is hard. So it was a four pronged approach. And he is the mind behind that and has done just spectacularly.


Justin Grammens  28:35  

That's exciting year right Hour of Code. I remember that years ago. And that's sort of how I think how it sort of flew across my radar, the company that I was working for we we tried to bring in actually, parents could bring in kids for the day. And we we were a software company, we brought enterprise backup software and security software. But the whole point of that day was was really to sort of not so much show about what we do, but really what it's like to work at a software company, right. And so we remember utilizing a lot of the tools like Scratch, for example, is a very, very great tool for kids to get started within this space. I mean, is there is there any evidence probably outside of because it wasn't your day to day at all anymore, but I'm just I'm wondering if there's something there for artificial intelligence or something there for data science, it may be beyond K 12. But I wonder if there were any of those talks or Acer underpinnings that maybe you could see from the outside that could apply?


Mona Akmal  29:27  

Yeah, so I'm pretty sure that code.org now has a bunch of courses that introduce kids to machine learning applications. They introduce kids to cloud computing, and what does it mean to work in cloud infrastructure? Because as we've discussed straight, so much of machine learning is about knowing how to work in a cloud native environment. Being able to understand how data pipelines were being able to figure out how to clean Data enough that you can do something interesting with it, right? So to me being a well rounded, applied AI person have to have a pretty solid understanding of all of those things. And finally, to know how to work with your engineering team so that you are effective at deploying your models, training your models, understanding when your models are drifting, and so on, and so forth. So understanding the full lifecycle, and I believe code.org has courses for cloud computing, Intro to Computer Vision, Intro to some machine learning applications as well.


Justin Grammens  30:35  

Nate, yeah, for sure. I'll be sure to put links off to that as well, for our listeners to get it get a chance to see what's going on there. Yeah. And so then let's move more into like somebody. So that's k 12. There's so there's somebody now that's coming out of college, what like what what would you suggest somebody that's coming out of school, even courses that they take, or things that they should understand, I guess, as or maybe just getting into the field?


Mona Akmal  30:55  

Yeah, so a few things I would say is you are, you know, just graduating and coming out and entering the workforce. It is important to work in a team where there is someone else that you can learn from right? Owner, that is a really strong engineering team that knows how to do data infrastructure and data pipelines and machine learning infrastructure, or it has a really strong science team. And selecting the company that you're going to work is really important, bigger company, or you're not the only person because you still need to learn you need a mentor. Second, don't pick a giant company, I see this mistake over and over again, where kids optimize or salary, your first 10 years, you should not be optimizing for salary 40 on you should start caring about, you know, minting money and bringing cash in, that's when you turn into a cash cow. A lot of sports are good, they're learning. So if I was 20, and or 22, when coming out of school, I would pick a hyper growth startup that had a really fantastic eblasts engineering team. That's super important because a lot of startups have garbage engineering teams. And if you are not surrounded by a plus b, boy, you're not going to learn and you're not going to become any plus person. Second thing I would suggest is all the areas in the lifecycle of an AI application, you should have a sound knowledge of all of them and the bar to find the scariest usually spend the most time. Because if you're not a well rounded person, by the time you hit 30, it's just going to start to reflect right? Because all the advice I'm giving is actually about optimizing your lifetime career arc, as opposed to going up really fast and then plateauing for the rest of your life after 30. Right? So I would strongly suggest understanding the entire lifecycle of an AI application. That means understanding how does data get ingested, if it's a massive data set, certain techniques work and other techniques don't work. So really understanding at that level of depth. How do you deploy a model, instead of just focusing on here's my Jupyter notebook, and here are the three packages that I care about, and someone did the work to clean up the data, and someone's doing the work to productize my model and put it in production. That's the wrong way to do science. And then the last thing I would suggest, and I'm giving you a slightly contrarian advice, instead of saying, take this course read this book, is, understand the business that you operate in my biggest pet thing with science seems is they build models that don't move the business forward, they might provide some amazing insight that is technically true, and will help them write a paper. But it is such an obvious and common sense insight that it just simply does not add any value to a business user. So really take the time to understand how to ask good questions from your business stakeholders, because at the end of the day, guess who's paying your salary? It's the business, it's not the publication that your paper is gonna show up. Right? So if you don't understand the business that you are in, and what business problem you are trying to solve, and what the parameters of that business problem are, you are not going to be able to come up with a model or a solution that is insightful or perceived as a solution by your business stakeholders, and then you're gonna be super frustrated.


Justin Grammens  34:31  

Frustration is never good, for sure. At least for for a long term career. No, I love that advice on I mean, I've been doing software development for 25 years or so. And I think yeah, for me, all the teams that I've done the best at was when I was not the smartest person in the room. Right? That's where I've learned the most. And it's interesting, a little bit off on a tangent here, but I think COVID And this whole virtual work thing is really it's really going to impact that because, I mean, I learned a lot by saying sitting next to somebody, right and being mentored through that now that somebody could be halfway around the world, but I think it's a different experience. And I think I'd be I'm kind of scared a little bit for people that are coming out of school and are purely virtual. And that's all they want to be as virtual. Because there's, there's that piece to it, where I went in and I grew so much, just by actually sitting next to and working elbow to elbow with a lot of people totally missed him so much.


Mona Akmal  35:24  

This is one of those things again, Justin, that you know, you can find me saying this over and over again, things you find scary are the things you need to be doing. Working with people face to face is one of those things, we've all forgotten how to do it, right? Because social skills are just muscles when you don't use them, they atrophy. But guess what, when you start using them and start building them up again, and I think all of us need to put in a little bit more effort to relearn what it means to be in a workplace. Yeah, we're never going to be 100% in the office again, because that's just stupid, right? It's counterproductive. But we'll get yourself to get out of your sweatpants and go and go to work and actually sit next to a person and learn from them. I hear this argument over and over again that you know, my productivity takes a hit when I'm not I when I'm in the office. No, it doesn't. It's a different type of productivity. Right? I'm 100%. Online.


Justin Grammens  36:25  

Yeah, for sure. For sure. Want to how do people get a hold of you? Is LinkedIn the best place? Yeah, and usually


Mona Akmal  36:31  

pretty responsive on LinkedIn. So just look up Mona Akmal on LinkedIn, or just email me at mona@falcon.ai. It's Falcon with key and we couldn't afford to see. So should be pretty easy to remember.


Justin Grammens  36:46  

For sure, for sure. Like I say, I'll be sure to put all that stuff as part of the liner notes on the site. Was there anything else that you wanted to mention today, at all that we didn't, we didn't cover I usually sort of leave that out at the end here for for guests who are on the show.


Mona Akmal  36:59  

I guess the last and parting thought I would share is, you know, AI is one of those fields where everyone thinks it's the leading edge right of technology. If you are in this space, please it is your job to demystify what it does and does not do scamming people by saying AI is going to solve all these problems. It's just like it doesn't serve us as an industry. So be a truth seeker and be a truth speaker about what AI is and what AI is not. So that we can all sort of work off of common understanding and common language instead of 99% of the world's population thinking that AI is Terminator, and the 1%. No, no. Right?


Justin Grammens  37:46  

Well as words, for sure, for sure. Yeah, make sure we're all level set the guards to what the capabilities are. We will all be not as frustrated in the future, as we mentioned earlier. So thank you, Mona, for your time so much today. I look forward to having you on the program in the future. And good luck with Falcon all the work that you and your team are doing. They're going to be exciting to watch you and the company progress over the coming years. But I know you have a lot of great things in store based on your history and all the amazing things that you've done. So thank you for all that. Thanks for being on the show.


Mona Akmal  38:14  

Thank you so much. Yes, it was a pleasure.


AI Announcer  38:18  

You've listened to another episode of the conversations on applied AI podcast. We hope you are eager to learn more about applying artificial intelligence and deep learning within your organization. You can visit us at applied ai.mn To keep up to date on our events and connect with our amazing community. Please don't hesitate to reach out to Justin at applied ai.mn If you are interested in participating in a future episode. Thank you for listening


People on this episode