camilla-pang-2020-06-17.mp3

Camilla Pang: [00:00:00] It's okay to link things that don't make sense. And if you find the links between them and they make sense to you, even if you like. Well, is that even useful? Yes, probably. It's been able to not judge yourself by thinking weirdly. That's probably one of the things, because when I was little, I made these notes. They made sense me. They made no sense to everyone else. But when you get older or when you as time goes, you might not make sense to me. So it's doing, I mean, cheesy, but staying true to yourself and to what you what your vision is.

Harpreet Sahota: [00:00:45] What's up, everyone? Welcome to another episode of the artist Data Science. Be sure to follow the show on Instagram @theofDataScience and on Twitter at @ArtistsOfData. I'll be sharing awesome tips and wisdom on Data science, as well as clips from the show. Join the free open Mastermind Slack channel by going to bitly.com/artistofDatascience, where I'll keep you updated on bi weekly open office hours. I'll be hosting for the community. I'm your host Harpreet Sahota. Let's ride this beat out into another awesome episode. And don't forget to subscribe, rate, and review the show.

Harpreet Sahota: [00:01:32] Our guest today is a postdoctoral scientist specializing in translational bioinformatics. She's earned a bachelors in biochemistry from the University of Bristol in a PHD in structural, chemical and computational biology from the University College of London. At the age of eight, she was diagnosed with autism spectrum disorder and struggled to understand the world around her and the way people were. Desperate for a solution, She asked her mother if there isn't a structural manual for humans that she can consult and upon learning, there is no such blueprint to life she began to create her own. This Blueprint, culminated in a book where she dismantles our obscure social customs and identifies what it really means to be a human, using her unique expertise in a language she knows best science. Her book is an original and incisive exploration of human nature and the strangeness of our social norms. Written from the outside, looking in her unique perspective of the world teaches us so much about ourselves, about who we are and why we do the things we do. And there's a fascinating guide on how to lead a more connected, happier life. So please help me in welcoming our guest today, author of Explaining Humans: What Science Can Teach US About Life, Dr. Camilla Pang. Dr. Pang thank you so much for taking time at your schedule to be here today. I really, really, really appreciate having you here.

Camilla Pang: [00:02:52] Yeah. Thank you for having me on here. I'm excited to discuss this. I really enjoyed this podcast, so. Yeah.

Harpreet Sahota: [00:02:59] Thank you. Thank you. I'm glad you enjoy it. So I'd love to hear more about your journey. So if you could just talk to us about your journey, how'd you develop and cultivate this interest in science? What were some of the struggles you had to overcome on your path to getting where you are today? And how did you deal with them? How did you overcome them?

Camilla Pang: [00:03:15] So basically, there's too many questions in one question. And so ironically, that is open ended questions for I find really hard. And I have this at the start of every podcast. And it's you know, I'm not even going to hide it, I think we should include this, because when I try and process something, I find it really hard to pick the details, which I need to say. And it's always been like that, like fun. I was little. You have no filter when you have autism, and that means basically taking everything literally and you don't know where to start. You're kind of marooned in the middle of this probabilistic landscape and you just don't really know to do. And so then when you try and communicate with people, you somehow don't know how. It's really weird. And for me, I've tried to make sense of human behavior and how to connect even from such a young age as four. And I needed something tangible and concrete to hold on to. And I read books and the things that I kind of, you know, clasped onto was science and math, because when I read it, I thought, yeah, that makes sense. And when nothing else makes sense, you're like, well, this is the substrate that I'm going to form my life around. And that came to me writing even highlighting science books with pens, copying bits out. And then when the science books in question can describe everything I used to stitch notes together and that formed the basis of the book Explaining humans.

Harpreet Sahota: [00:04:32] So that's pretty interesting because I really enjoyed going through your book and you've got some awesome notes and drawings in there as well. So I cant wait to dig deeper in and talk about that but before we get into that, I was wondering if maybe you could pick your brain a little bit on what you think the next big thing in machine learning is going to be, you know, in the next two to five years,

Camilla Pang: [00:04:52] Two to five years? That's quite interesting. I think I mean, to put a timeframe on it, to be honest, I feel like the attitudes we have generally in machine that is to replicate, you know, the kind of logic within the human brain and that's going to get more accurate we're going to have more data. And I think there's going to be a point where we realized that this element of precision that we're all aspiring to have even as humans, is not going to be as effective as we think it's going to be. So, for example, there needs to be more nuanced and there's going to be question in what we know, because everything about A.I. is based on what we know and what we want to kind of predict. Whereas the human mind is a lot that we don't know. And that still catches us off guard. And that's the same for everything else in life, new complex adaptive systems. And so the thing that I'd like to be reading about or be looking at is more things like an agent based modeling and modeling these complex adaptive systems, not assuming that one rule fits all or one logic fits all. So I talk about it for quite a while. I think we're definitely going to appreciate the limitations of very rigid algorithms and try and incorporate more flexibility in absorbing and making the most of chaos as opposed to try and kind of spread it to a side. So that's what I'm hoping, I know that's quite a vague answer but. 

Harpreet Sahota: [00:06:08] Thats definitely , very, very interesting. So, you know, with this kind of vision you have for what it looks like in the next two to five years. What do you think would be the biggest positive impact on society?

Camilla Pang: [00:06:18] Oh, its like many things. It depends how is used, isn't it? So you could make this amazing tool. Then it's like me saying, oh, yeah, I'm gonna make a clone of a super human, which, quite frankly, is what we're kind of trying to do with A.I . But then. What is a superhuman do? So I'd like to feel. I'd like to think that A.I. is used in a way which can make more sustainable solutions and not kind of accelerate this whole capitalism that we're seeing. It's more for the greater good. Basically, you know, I could say I'd like to see applied to things like climate change. I'd like for it to be able to predict things are more to do with mental health. That there's my subjective causes. But I'm sure there are many other things. But before we get on to making the most, of A.I we first need to make the most out of human minds.

Harpreet Sahota: [00:07:04] Very, very interesting point. And I 100 percent agree with that. Yeah. So just kind of continuing on this wave here, trying to pick your brain about the future. What do you think would be scariest applications of machine learning in the next two to five years?

Camilla Pang: [00:07:18] I feel like of a similar point the things that scare me most about this emerging technology is, you know, humans started to see it as the kind of almighty consciousness. They try and believe A.I. more than they do with, you know, a human. So I feel like it's how we interpret it and knowing its limitations. The thing that scares me a little bit is people assume that because it's not human, therefore it is the, you know, the right answer to everything. You know, at number 42, Hitchhiker's Guide to the Galaxy. So basically, the thing I'm scared of is people trusting A.I. more than they do humans.

Harpreet Sahota: [00:07:51] So I'm wondering, in the vision of the future that you have or what do you people separate the great Data scientists from the merely good ones?

Camilla Pang: [00:07:59] Well, the thing is, we have lots of different shapes of data scientists. And I've actually noticed that there's actually not one type of data scientist that is correct. And it's acknowledging this and making the most of the soft skills that you have. I mean, I could say that I need to code and do an all nighter every night. But then what would you achieve? You'll achieve you might you know, there's many different things that make a data scientist other than coding. And this is something only I've recently learned because I've only been in bioinformatics for I don't know I say two or three years, a little more than that many five now. So I'm still quite new to it. And I did a personality test at work and I was actually quite scared to show my boss the results because it was not Data scientist shaped. It wasn't as technical or logical or organizational as one would, you know, stereotype. And then you're like, yeah, that's fine, because we need different types of data scientists. So it is really nice to hear you know. So mine was more emotional and conceptual. And I feel like acknowledging the different shapes and to improve A.I. we need to appreciate the nuances in people. For a start, and not everyone is to be tunnel vision and doing one skill. So it's been adaptable and knowing what you can offer.

Harpreet Sahota: [00:09:13] I absolutely love that response. I think that's 100 percent true, that the ability to just have different shapes of data science. Sorry, I go on go through your example.

Camilla Pang: [00:09:21] I know. I've just seen about my strengths. I'm learning how to code like I'm not a python wizard. I try and learn it so that I can try and apply it to the problems that I'm faced with at work. And just for general, you know, funnzies. But what I'm really good at and I think this requires an element of competence, is being able to envision and look at how algorithms are simulated in real life, but also in practice.

Harpreet Sahota: [00:09:47] And if you could talk to us about the terms neurotypical and neurodiverse. Would you mind defining these terms for our audience?

Camilla Pang: [00:09:54] Yes, of course. I've had this question a couple of times, and I think every time I say it, the answer's only slightly bit different. And I think it's because it's an evolving term. So I guess there's two definitions of it. If someone is neurodiverse or neurodivergent, you I mean, everyone's neurodivergent. I mean, it's like we genetically divergent because we're not all exactly the same. But when it comes to neurodiversity and having attributed, I guess, diagnosis, so you are so different enough to not be able to function in the normal. I guess everyday life, it hinders everyday life, basically. For me, I have a form of autism. I have autism spectrum disorder. I also have ADHD. So Attention Deficit Hyperactivity Disorder and generalized anxiety disorder. I know that sounds like a lot of acronyms, but what I really want to highlight is that everyone is nerodiverse to the point in which it hinders everyday life. You will have different parameters of your brain which are altered to such an extent that you can't even focus on tiny little sayings. Or when you say something is outside, is that outside what is expected of you? And I think to be able to be neurodivergent and own it and know your own shape is very brave because a lot of the time when you are neurodiverse diagnosis or not, because it's actually very hard to get a diagnosis, especially in adulthood, you will feel squished and you can't be yourself. So tha is also something that I would attribute to someone who is neurodivergent is one they like. I don't understand my shape, even if I did. I know no one else would.

Harpreet Sahota: [00:11:20] Yeah, definitely. Very interesting. And thank you for defining that for us.

Camilla Pang: [00:11:23] I tried, I'm sure someone will be like oh no or something else. But this is the thing is divergent.

Harpreet Sahota: [00:11:29] Yeah, that's the thing. Right. Like some of these terms, like it's how you internalize them and how you've come to understand them. I mean, there's always more than one right answer, I think. I'd like to get deeper into your book. So I think it's fascinating in your book how you draw parallels between machine learning and human cognition, especially in terms of decision making. You talk about two things that I thought were pretty fascinating, among many other things, and they were thinking in boxes and thinking in trees. Can you talked to us about what that means. What does it mean to think in boxes and what does it mean to think in trees?

Camilla Pang: [00:12:04] So basically, simply put, I tried to attribute these different types of I guess I call Data structures or like these categories vs. an evolving, branched stream of thoughts as two opposite ends of the spectrum. And we all have them. We all do both. But to be boxing computer box thinking is there's no room for error. You're like ticking to the time intervals you like. I got 22 minutes to do this and it's great because it means that you can be very efficient and you are the fidelity of thought of you doing X, Y, Z is, you know, very definite. It's very good for acting things there and then. But sometimes what happens is I used to be like this through and through because that was how I implement the structure of my life. I didn't know what else to do. I felt that there was only one way. And I thought, OK. Does that mean I just do this can i understand it? And most of the time it's knowing what you like and liking what you know and with box thinking. It's categorical in that regard. And that is there's only a few alternatives and you can't really see the solutions in between. You get quite stuck with to think in trees just like all it. And it's more of a branched, I guess, probabilistic landscape. And I talk about this more and later chapters in the book, but primarily it's about being able to acknowledge the different branches of fates can arise from moment in time and be able to reassure yourself that there are, you don't have to put all your eggs in one branch and you can separate them out and be like. I could do this. We could do this. And if this doesn't happen, then let's do this. So it's kind of a reassurance based on experience that was helps because in, you know, the eventualities. But sometimes you get stuck in a rut and think so much in trees that you don't end up doing anything because every movement that you make could be a wrong move. So this is why boxing can also comes in handy, because it helps you to coalesce your streams of thought into action. And I think this is one thing that we're missing in lockdown is these boxes that we're used to having. I certainly am, anyway.

Harpreet Sahota: [00:13:58] So sounds like there's not really one mode of thinking that's better than the other. Does it really depend on what it is you're trying to solve, problem wise. What it is that you're thinking about.

Camilla Pang: [00:14:07] Yes, a bit of both. And this is ironically, this is by acknowledging that we can do both. Is tree thinking in itself? Because it does depend. I've tried to assimilate the extremes of both. This quarantine, I've not done box thinking at all because I actually find it hard without my normal environment. And so I've actually been less productive because I'm like, well, I could do this. And then you're stuck in the middle and you're looking at your knee for two hours. That's not what you want. You want to be able to make a decision there and then. And this is where you need to notice what you're doing. And then how can you go to the other one to get things done? So it does depend on who you're dealing with. For example, if you're trying to think of flexible solutions and it's good to know what you know, but to kind of connect the boxes and to feel like you're not on a cliff edge of decision thinking in this branch like none, I can offer a lot of reassurance and also alternative solutions that you wouldn't otherwise thought of. If you just stepped back a little bit. 

Harpreet Sahota: [00:14:59] And you mentioned that people tend to be stuck in a box thinking type of mode, why do you think that is? That most people are stuck in a box thinking type of mode.

Camilla Pang: [00:15:09] It's a creation that they built for them to survive. And for them to function almost like an algorithmic module. OK. This is the module. This is the function. Also emotionally what that means for a human, is you learn something and you're like, well, it worked so far. The model might be wrong. You start to question other alternatives and try and squeeze them into the vision of what the world should be. I mean, that is one of two ways really. But it can be quite limiting and you can get very bored easily. So it's good that you get bored easily because it means that naturally you are quite a tree thinker. You know that there's always more to do. But it's based on fear ultimately, because obviously you've invested in this box like vision or box is like visually pixels. And you're like well, how else do I see the world?

Harpreet Sahota: [00:15:49] And for those of us who are able to think in trees, what can we do to get beyond the first branch of that tree?

Camilla Pang: [00:15:57] For me, much like...We're going to refer to Data science now, because, you know. This is a term that I recently learned could make jittering. So basically, when you do a visualization, you've got many points for one point, basically, and you want to see them all and so you kind of just do this random scattering about points that you can kind of acknowledge that they're in the same space. So what I try and do is do that. Everything that I'm doing, if I haven't got routine, set up, I don't know where to start. Then why do it? Okay. Do I write, Okay. I could pick up a pen or I can do this to this. Maybe I could do this. What's related to this? And I think that randomization is something that I find. Useful in defining what I actually need to do there and then. That's because and it cushions me around what I'm trying to do and what I want to get from it. It's hard to know where to start. Which is why I guess procrastination does help me in that regard to help centers me where I need to be. But honestly, make sure you put your phone away. That's another rabbit hole.

Harpreet Sahota: [00:16:50] Yeah. So you talk about learning to embrace errors and why it's super important to do that. So what can we do to start embracing errors in our own lives?

Camilla Pang: [00:17:01] I think it depends what people mean by errors. So for example, error and uselessness are merely just a byproduct of not attributing to a you know, an orchestrated utility is not good for this. Therefore, it's an error. But actually reshaping how what we think of what error is and can be quite useful, because an error in one context is a solution in the next it is depends on what your viewpoint is and when it comes to being neurodivergent you would have multiple different viewpoints in the space of a minute, which is why you A you always see the ultimate solution. But when that's useful enough, that takes you zooming in on it. And that can take confidence. So with errors, I know as many different ways of doing doing one thing. But to acknowledge errors are to acknowledged spaces in between the tree like thinking what bunches them together. I think by acknowledging what we define as noise isn't completely useless, but could be used in other contexts as being flexible with that. And I think back to your question earlier, to see the signal from the noise and to see the opportunities that can arise from it will take what is a very good feature of a data scientist, not have the tunnel vision So yeah. 

Harpreet Sahota: [00:18:10] And I thought it was really interesting how you mentioned it, like the knee jerk response to error is like a downfall of box thinking, where you kind of categorizing everything so if it's like, oh, that's not right. Immediately, just kind of given automatic response to that - did I understand that correctly.

Camilla Pang: [00:18:26] Yeah. Basically, it's highlighting. Yeah. That resourcefulness is not useful now in this context. But why would it be useful in the next. We don't know dis it that we won't acknowledge it's there and question why it's there. We'll get on with this. But it's not been like oh that's wrong. It's like saying to a person like people say to me, oh no, your, you know, your, your existence is wrong. Huh! Cheers. No, it's not. It's just different to yours. So I feel like I don't want to extrapolate from actually from a data point to a human. That's the same attitude.

Harpreet Sahota: [00:18:54] It's really interesting.

Harpreet Sahota: [00:19:02] Are you an aspiring Data scientist struggling to break into the field or then check out DSDJ.co/artists to reserve your spot for a free informational webinar on how you can break into the field thats going to be filled with amazing tips that are specifically designed to help you land your first job? Check it out. DSDJ.CO/artists.

Harpreet Sahota: [00:19:27] But it's really interesting how we're making connections between proteins and personality and interpersonal relationships. So what do proteins have to do with personality and interpersonal relationships?

Camilla Pang: [00:19:39] So basically, just to bring it back to the book is why I used proteins as a means to model humans. Whereas when I was little, a lot of people use attribute personalities and to learn from the personalities of the teddy bears and what that means for humans. And they made characters out of the things I knew that resonated with them. And they connected with I didn't reconnect with those or people and I knew that I connected with science. And back to the chapter I mention that, quote, Proteins are like humans or humans are like proteins. It was a way of me being able to model this dynamic behavior I saw on a football match, as one does. And I realized that this dynamic behavior could be modelled. So I knew that a lot of people that they moved in cliques and they moved independently. But then one person was different in one context to another. And I really love this disparity between what it meant to function in one environment and then be able to adapt to another. So, for example, it provides a good model of human behavior because it enables humans to be considered in different adaptive contexts as opposed just having one function. That's why I liked it.

Harpreet Sahota: [00:20:50] So how could we use this understanding of proteins to be better colleagues and better teammates at work?

Camilla Pang: [00:20:57] Ok. So for example, when I am so when  wrote the chapter, it was quite...I wrote all this book before I was 20. So at the moment I hadn't actually been in work. But now thinking of it, what I did is try to tribute these different types of proteins and their personalities or their time to function and their responsiveness in the cell and what they did. You got the receptor proteins, you adaptor proteins, you kinase and your nuclear protein - I'm sure there's many others. But what I like most about this protein model was to do with the fact that you have different layers of structure. You got the sequence and you've got the primary structure with the sequence of amino acids and then you've got it folds in on itself to create this module of evolution that is then into wound with other units of the evolution to kind of help characterize the different modes of function. And when it came to humans, I thought, well, what's basically the same as us? We're ultimately determined by our genetic sequence, along with our environment, among many other factors that make us human and make us behave differently in different contexts. So for me, I was like, well, I saw this. I was like, this is great. I was to try and make it a bit more accessible to people in the book.

Camilla Pang: [00:22:02] I like to kind of parallel the worlds of proteins to a well known psychological tests. Called the Myers Briggs classification and just make it a little bit more human? Because loosely, when people see proteins, you don't see personalities. So I tried to parallel that. But I think one of the limitations of the Myers Briggs is that you limited to a 4D, you know, four letter metric and then thats it where the protein could be many different things in many different contexts. I mean, you could have a tertiary structure, which is basically like a blob of, you know, can coded by one genetic sequence. That can then interact with another kind of protein and tertiary structure where the blob to have a different function. So therefore, it depends a little bit more on what the cell needs. And so this is something that thankfully I'm in a job that I think that makes the most of the different sides of people. And for example, if I'm not Data scientist shaped that, that's OK. So it's not just getting to know people. It's about knowing the shape, the different sides of people and what they can offer in a team. And also, it goes both ways. Is being able to open up and be like, actually, I'm good at this also. And speaking up so that you can shine, basically.

Harpreet Sahota: [00:23:09] What tips do you have for, let's say a Data scientist in a team environment who might be scared of looking like they don't know something, or maybe scared to admit that they don't know everything, but they do want to openly communicate that to their teammates. Do you have any tips for them on how to overcome this type of fear?

Camilla Pang: [00:23:27] That's a really good question. And this is something that we often sometimes ask myself as well. Some of my friends ask me, and it's knowing your own shape and admitting that because you don't feel like you are on top of the main skill that kind of pinned you down to the job and be like, yes, I coded this and I'm going to deliver this. That doesn't mean that I mean it means you've done the job, but not all of it. So I feel like I'm hoping that in the next two to five years, there's gonna be an increased awareness that there's gonna be many different Data science shapes. And from that, people are going to really that they have more to offer and that they can be that shape. So when you are feeling like you have more to offer but you don't want to open up, ask yourself. Okay. Brilliant if I'm great at this how is this gonna be useful. It is not useful then. Where is it useful for. And then before you know it, you can kind of route down to how it is useful and work and might even just be a collaboration with other labs. Well, that doesn't involve coding. No. But it does involve being able to communicate between different sides of science and bringing it altogether. And surely that's the philosophy of Data science anyway, isn't just being at the computer. So there's many different sides of data scientists that cover the basic fundamentals of being human in my eyes, because you have to be a scientist and an artist to have a vision and see how things connect. And also to communicate with experimentalists and higher order structures, for us to integrate the main question to actually this is something I tell myself and I feel reassured because even though I probably want to spend more time coding and I try. I know that there are other sides to my personality that also make me a great data scientists such as, you know, it's what energizes you if you if you're really not a good fit for the job and something you have to question. But ultimately, the more we get to know Data science, the more we realize that we're all scientists at the end of the day because it requires a science and an art.

Harpreet Sahota: [00:25:16] I love that response about the science and art being blended together. I like how you talked about the principles of gradient descent in order to identify and prioritize our goals. So for, you know, the non Data scientists, listeners in our audience, would you mind just quickly describing what gradient descent is in layman's terms.

Camilla Pang: [00:25:37] In layman's terms, yeah. For me, it means trial and error. And the main bit of trial and error is knowing which bits of the kind of landscape of solutions is better for the short term. So local kind of solutions or local optimum. And then you've got the kind of bigger picture, which is a little bit vague, but we kind of aspire to it. And we know that there's a better solution out there. I can't quite touch it yet. So we kind of keep exploring. And then we kind of come to a point where like hold on a minute, we can't get much better than this point of convergence. And so this is the global kind of solution of this scenario. And we could even argue that this is the case for lockdown. We're trying to find these global solutions on every day local actions. And which one's optimal for today might not be optimal for two months time. And so we're in this constant battle of local versus global solution. And to be able to navigate this landscape, you have to iterate and block this work and then you kind of be able to kind of be oh, no, that wasn't good. I want to backtrack. And it's been able to execute trial and error in a way to find these global solutions. So I hope that's kind of clear enough.

Harpreet Sahota: [00:26:47] Yeah, that's great, because we definitely need to have, like, the way you interpret it, the way you understand it now in order for us to kind of explore how we use this to solve or rather prioritize and identify goals. So I'd love this analogy. So given, you know, the definition of gradient descent has provided us. How can we use that to or how can we use that framework, that mental model in order to help us find our path to prioritize and identify our goals?

Camilla Pang: [00:27:13] I feel like people do this anyway. So this is something that I'm not proposing a new algorithm for humans to implement. I mean, people do this anyway. We're just trying to make machines implement them. So on a scale which isn't judged, because if we were to humanize algorithm, it's already on a computer, we'd call it an anxiety attack because quite frankly, what we're trying to do is simulate all these different solutions and for it to go up and down all these different, you know, highs and lows and to find a solution. So this is where I realized I was doing a gradient descent whenever I have a meltdown, because it was like oh no no she's being silly and, you know, I'm like crying. And my head, my head is really hurt and I've got my hands on my ears. I'm doing a gradient descent, but accelerated to the point where I need to find a solution of convergence because there are times and where your head is spinning. And this can often happen in a meltdown. This is where I don't mind having meltdowns. I know it often enables me to reach a point of convergence in a trial and error just through, you know, dynamic simulation to be okay. I need to do this now. So it's basically something that everyone can use, but it's discriminated against on a kind of accelerated scale because it's basically you are just doing it everyday. Trial and error or you're doing it in the form of an anxiety attack. But people don't realize how powerful anxiety attacks are, because whenever I have this, I call them storms in my head.I feel a lot clearer after. Because I'm like okay. I know whats most important to me.

Harpreet Sahota: [00:28:37] Yes. Very interesting. Really laid it out in your book. No hope in people go and get the book after this because they can make a lot of great analogies using, you know, math and statistics to pretty much the human mind. I think it's really fascinating. One of them i really liked was you talked about probability and empathy using Bayes Theorem. There is a method that's pretty well put. So I'll take it for granted that my audience knows. The difference between Bayesians and frequentists and they understand what Bayes Theorem is. So how can we use Bayes Theorem for empathy and managing the relationships that we have with ourselves?

Camilla Pang: [00:29:12] So basically, I originally used Bayes Theorem as a way of being able to not just take things at face value, because when you were literally that is when you have autism and especially in my form of Autism especially I take, things literally which is as they are. And that is great. But also it means that not many things are in context. And so you're trying to make this context around them. Okay. They're angry at me. And I don't know why they're angry at me, should I be angry or should I be happy? Why they upset? Oh, no, I got them this, I to do this. So Bayes Theorem, I use it as a way of being able to simulate or kind of contextualize the words, the actions and the characters of peoples that I know how to respond best and to make them happy. So if, you know, obviously it takes a while to get to know someone. And what I also described in this chapter that use Bayes Theorem is that getting to know a person in situations is like cellular evolution. Your a stem cell at first when you kind of could be anything, and so therefore you don't necessarily specialize. But as you go, you see this kind of outward hierarchal structure of cells that are a little bit more specialized each time. And it's an absolute beautiful diagram, it's actually one of my favorites actually. It's hemapheresis the kind of differentiation of blood cells and the immune system cells. And from this, you can see the parents cell of which each cell. So it's like, oh, you keep back in track. You kind of come to the starting point. And what I tried to do from this is I saw Bayes Theorem in this. This wouldn't have happened if this cell didn't occur before. And so I use both cellular evolution and Bayes Theorem as a way of being able to simulate the events that happened beforehand. And the data gathered, you know, of this person, for example, to know more about what makes them tick and more about what makes them happy or when they're upset. How can I make them feel better based on what's happened before?

Harpreet Sahota: [00:31:02] Yeah, it's definitely very interesting the way you laid it out in the book. so, again, guys if your listening to this you have to get this book. It is super cool. So what is a neural network and how can it teach us about ourselves or what can it teach us about ourselves?

Camilla Pang: [00:31:15] So that's an open ended question. And my mind is spinning because I'm like, well, people do neural networks the time they basically are a neural network. But massive and interlinked. And it's about this process of combining all these inputs and assembling them in all these different combinations so that you can kind of make a decision. And if it doesn't work, like, oh, no, go back. So it's ability to assemble the input information to come up with all these different combinations of which ones work best. And then from that have response leaps of feedback loop that goes. Whether it's good or bad over time, there's lots of different brands which you can do this particular event or situation. Humans do this all the time. It's our ability to take in and process information and then make it and make a decision on how to act. And from this act is then feed back into the original, I guess, data that you sense. So for me, that's what it is. I think there are other ways which people describe as a Perceptron, which mean that it's not this small neural network, but that's the basis foundation of being able to - you know, that's the parallel between human and computer. You sense. You process. You output. Then you feedback.

Harpreet Sahota: [00:32:17] That's what I really enjoyed about the book was the fact that actually, you know, I think at a fundamental level, like machine learning, these decision systems, they are created to model kind of, in a way, human decision making processes. I just like the way that your book really makes it explicit and then provide some real world examples from, you know, your own life the way you think about things. So I thank you for that. So, you know, you're somebody who is a practitioner and you understand the subject of machine learning and data science. I'm wondering how you view Data science and machine learning. Do you view as art or as a science?

Camilla Pang: [00:32:52] A bit both, just to put it bluntly. I think people learn it as a science, but then they realize that things go wrong and then they're like, oh, no, why is it go wrong? Because we assume that the machine has some more nuanced than anticipated and then trying to get the code right. It's also like talking to a human that doesn't give you any feedback. You're trying to decipher why the code's going wrong. So when it comes to the everyday grind of Data science, it literally is an art that takes a lot of patience, still trying to work on. But also to be able to curate what reality is on the computer so that you can model it. I mean, that takes an artist. but you have to see things in between the lines and the know how to model them on the lines effectively both.

Harpreet Sahota: [00:33:30] And how does the creative process manifest itself in Data science or in science in general, rather. 

Camilla Pang: [00:33:35] I'm basically writing another book on this? So at the moment my head is spinning. But thats cool because I was trying to distill it down to a very small sentence. I'm actually reading about that now. But as scientists, we're not. We're stereotypically, we're not meant to be creative we're meant to be rational. But actually, one of then I guess the mistake in stereotypes is that to be a scientist, you need to be very creative, need to think between the lines need to be able to envision and be resourceful in different contexts and make things and make theory turn into practice. And from that theory, that doesn't mean anything if it doesn't model anything in real life. Both art and science are just an appreciation of the world and being able to feel like how can I model it in the most effective way possible. And effective? And that's where it mainly differs in science and art. Effective in science means objective, representative, logical. Art is everything else in between. It's the noise. It's the nuance everything that you probably wouldn't want to put down because you worry that it's subjective. But actually, this is where they're very much complementary, but also it's about the attention to detail. So what level of granularity are you able to pick up on to make up your picture? And from that. This is something that I worry about when I'm doing my own simulations. I worry they're not going to be good enough. I don't know what that means. All I know is that. Does my intuition of a Data scientist with this specific, detailed profile. Is this going to be what they envision? It might be. It might not be. So there's no one solution to the problem. And I feel like that's one of the things that a lot of scientists need to know. Also, myself included, because we can have imposter syndrome that we know that everyone's got a different type of vision.

Harpreet Sahota: [00:35:11] Thats definitely very important to keep mind - that different Data scientist approaching a particular problem statement will have their own unique way of making progress in solving that problem statement. But as long as they're able to justify every step they're doing along the way, I think that methodology, the methodological part of it is like the science. Everything else, like you mentioned, like the noise, the grey stuff in between. That's the art and play. So on throughout the book and on your Instagram as well, I see this post. You've got your books and your notes on Instagram, and then you have this drawing structure book explaining humans and I really like that note-taking style. I think a lot of our audience would benefit from you kind of taking us through a process for taking and making notes. Would you mind sharing your note, taking process with us?

Camilla Pang: [00:35:56] Yes, of course. Whenever I make note, I've noticed that having books that are OK, these my notes today and they're not very structured and to be structured, you need everything to be make sense in your head. I write to make sense. So you've got chaos. You've got all these different weird scribble marks. You like a map or an atlas of the problem. It's me trying to boil down the different elements of dimensionality into something that is more coherent, such as a paragraph that much like Data science takes a lot of wrangling and a lot of processing to make sense of what you envision in your head. So for me, I like to - and this is a form of artwork. You see the links and the bonds between the words and each angle of the paper. In my pile means something. It's a message much like the terms and DNA. You have all these different encodings of which paper, and what angle means what? And for me, obviously, if the wind blows, then, OK, I'm going to reassemble it. It's a narrative that I'm very sensitive to. I like to physically see everything laid out in front of me. So, for example, it's its own message. And even when I'm at work, whenever I make to do lists, you see them and they're really messy. My boss is like maybe we should type them up. And I'm like, yeah, maybe but it just doesn't feel the same. So I'm constantly trying to be neater. But then the information I put out doesn't feel the same. So it's good to do a bit of both. And if it's messy, then if it makes sense to you, then that's fine. But yeah, it's basically like an artwork. It's very messy.

Harpreet Sahota: [00:37:26] Yeah. Really. I'm a big fan of like that, almost kind of like mind maps in a sense has kind of reminded me most of. But it really does like just looking at the drawings that you have for the book, its inspiration for me and how I want to think about taking notes. But also just it distilled everything down to a picture. It's really cool. So I think it's a lot to be learned from that, too. Thanks for sharing. Thank you. So a lot of data scientists, whether they're in their actual jobs or whether they're trying to break into the field. Projects are a part of what they do, and they may be feeling some type of hesitation or fear because they're wanting to make their project absolutely perfect before releasing it to the world, before releasing it to their boss or what have you. Do you have any tips for anyone who is stuck in this kind of. It must be perfect. Before I release it mindset.

Camilla Pang: [00:38:11] Yeah. So I'm a bit of a perfectionist and so I completely get it. It's good to question and humor yourself. What does perfect actually mean? And then your like actually what does it mean, does it mean it has to be this color. And then you be like nah they can't be that specific. And then you start to kind of reason with yourself as to what your definition of perfect means. And for example, it might be, you know, when you're working with, you know, experimentalists, working up scientists, everyone's vision of perfect is very different. And so it's knowing a bit more about your team and what they want. And even though it might not feel like the perfect solution for you, if it does the job, then that's fine. So it's yeah, I definitely feel like it's something that a lot of people battle with. I think communicating more of the expectations are, you know, that upon your role, what you need to do. If you know what the wiggle room is and you can kind of work around that. But to make it of a quality that your work justice, that is not a personal endeavor isn't that? So, it's not to get rid of perfectionism. Just making sure that when you do it, you're doing it yourself. Also for the needs of the team.

Harpreet Sahota: [00:39:10] Thank you for that. I think that's really valuable advice that our audience is going to benefit hearing. Thank you so much for that. So how are soft skills for a minute here? What are some soft skills that you think Data scientists are missing that are really going to help them excel in their careers and in their interpersonal relationships?

Camilla Pang: [00:39:30] So whoever coined the term soft skills they're wrong because soft skills are actually really hard work. Because the nuanced and they're context dependent. And no matter how friendly you are, you're always going to end up annoying someone. I think that's just a given. And that's actually quite nice to know because you, might end up catching on a bad day. But when it comes to soft skills, I feel like from my experience so far, for me it's a bit different because me, it's mainly being able to ease anxiety in myself so I can communicate effectively with my team. And it's also being able to communicate and making your team feel that they can come to you and ask you questions. So that's one thing I've learned is that people that don't have as many soft skills, people don't want to come and ask for help. Because they feel like they can't. Or they're on the box thinking, they're cliff edged by someone's judgment. But if you feel like, yes, OK. If you open with them and being able to not judge them based on them not knowing something, I think it's having that friendly banter. And to see them as a human, as a friend. And I'd like to do that with all my colleagues because I really I think that's really beneficial, because when you're an off day, you can kind of like talk to them and you can. I think that's the soft skill for me that's been most important is even now and most the time everyone's got their headphones in, which is great. But underneath all, you have no idea how anxious they're feeling when they're coding. So I think that's a soft skill that your need to teach yourself is being able to reassure yourself that even though everyone looks like they know what they're doing, they might not. They might be as stuck as you. So it's been open enough people to you can help each other, basically not judge each other because there's this whole kind of stigma. You have to be rational. You can't be emotional. You know, emotions are weak. They're a soft skill. We don't need them. But this is a form of toxic masculinity. We need to be able to make the most of the different sides of people to work effectively.

Harpreet Sahota: [00:41:19] What's up, artists? Be sure to join the free, open, Mastermind slack community by going to bitly.com/artistsofDatascience. It's a great environment for us to talk all things Data science, to learn together, to grow together. And I'll also keep you updated on the open biweekly office hours I'll be hosting for our community. Check out the show on Instagram at @theArtistsOfDataScience. Follow us on Twitter at @ArtistsOfData. Look forward to seeing you all there.

Harpreet Sahota: [00:41:49] 100 percent agree with you that soft skills are a bit of a misnomer because they are really the hardest skills. And I think at least from my perspective, they are the hard skills because they can't be taught. You learn through experience.

Camilla Pang: [00:42:04] You learn from experience. Yeah. And just to put in that further, in Chapter eleven, I talk about how to be polite or mainly about the kind of nuances of etiquette and how to model them. So there is there is some ways you can kind of benchmark whether you're doing it right or wrong. But ultimately, they can't be taught, which is my one of the conclusions of the chapter. But not all of them.

Harpreet Sahota: [00:42:25] And to your point, I think, like vulnerability is definitely a very important soft skill. I think once I sort of embrace that in my professional life, when I was just open about like, yeah, I don't fucking know what the answer to this thing is man. Give me time to look it up and research. And once I became okay with not knowing everything. Things just became so much more easier.

Camilla Pang: [00:42:46] Yeah, definitely. Which is where the protein model comes in because they're always evolving. Everyone's not just this one dimensional being. We were constantly evolving.

Harpreet Sahota: [00:42:54] So I mean, I.N.F.J. personality type on that Myers Brigg scale. What what protein would you say would best describe me. 

Camilla Pang: [00:43:04]  ooo you've got. I guess it's a nuclear protein, isn't it. So it's one with a guest on it or the nuclear membrane. If you're talking about that. It's someone -  so I.N.F.J. So what I've tried to do, even though you like, what's the point in using the protein model if you just map it to Myers Briggs, which is a very good question and one I ask myself many times throughout this podcasts. For example, if you were to equate the two would be a nuclear protein because you get along with it. I mean, you're receptive to it, but that new kind of doing your own thing quietly, but then you care about other, you know who you affect and how you make people feel. This is what cause I'm an INFJ as well so snap. This is the thing. It depends on the context. And I don't really like talk about it too much because it's I don't feel a lot like has much to offer. And people are ohh what protein am I. there's actually a quiz. I am saw back up now like ten years ago maybe about what protein am I. And for some reason I got Kinase, which I thought her hilarious, which because kinase would be a very extroverted, dynamic kind of like party animal and like, you know, spoiler alert, I'm not that. So when it comes INFJ, such as she knew it would be a nuclear protein.

Harpreet Sahota: [00:44:09] Thank you. Definitely. I'll look into that a little bit more. Never, never saw myself as being very nuclear. But I like the way you spelled it out in the book with the different mappings. That was really interesting.

Camilla Pang: [00:44:19] It depends what you call what context. I mean, it could be a Kinase when you feel really comfortable. So this is why I - this is why it changes.

Harpreet Sahota: [00:44:26] Yeah. So I was wondering if you could speak to your experience being a woman in STEM and if you have any advice or words of encouragement for the women in our audience who are breaking into STEM or maybe they're currently in STEM, might be facing, you know, any manner of adversities. Do you have any words of encouragement or advice for them?

Camilla Pang: [00:44:47] Do not try and hide your femininity because you feel like it will make you more logical. Doesn't work. I feel a lot of women try to mask their femininity because they're worried that they're going to be frowned upon or they're going to be silenced because ah yeah, I know thingy said that, you know, cause she's, you know, she's you know, that people will judge more. They're worried that people are going to judge us more because we wear lipstick, because we, you know, we like to wear perfume or we like to have all these things that make us feel good because it's from an emotional place and therefore we're less logical. I feel like a lot of women try and silence is part of themselves because it makes them feel less of a Data scientists, even though they feel better as a woman, that it shouldn't be mutually exclusive. You're just a person who feels who has their way about them and shouldn't have to sacrifice a part of themselves in order for them to be listened to. But this goes for other people in Data science [inaudible]. It's a two way thing. It's not just, oh, we need we can give women a voice and we can empower them all we like, but we can scream and shout as a per say. But what it takes is for people to listen, not judge based on what we look like. Based on what if we're having anxiety attack? Ah yeah she's having an anxiety attack yeah she's unreliable. That that's something I feel that could be helped a lot. I mean, I'm very, very lucky. I'm in a work environment where people know that's my nature, mainly attributed to the fact that I'm, you know, I'm autistic. A lot of women go for this inside and a lot of men, actually. But I think to be able to show it is something that takes alot of bravery and it requires a person to be empowered. Also that the work environment to be receptive and supportive and not silence based on the fact that someone has certain shape.

Harpreet Sahota: [00:46:21] And what can the Data community do to foster the inclusion of women in Data science and AI and STEM?

Camilla Pang: [00:46:30] I think it relates a bit more back to vulnerability, being able to talk a bit more about what you find difficult and open it up. It's actually quite well in my eyes its quite simple solution because is to be human. For me, I don't I mean, I shouldn't say this, but I find gender something. I see human as a human. I don't see them. Oh, yeah. You're a woman. Oh, you're a man. I'm not. Okay. You're person. Cool. You know? So I feel like a lot of people should go with that shit. I think a lot of them do. But it's just being self-conscious about what you portray and not knowing what people think of you.

Harpreet Sahota: [00:47:00] Last formal question before jumping to a quick lightning round here. And that is what's the one thing you want people to learn from this story?

Camilla Pang: [00:47:10] It's okay to link things that don't make sense. And if you find the links between them and they make sense to you, even if you like what, is that even useful? Yes, probably. It's been able to not judge yourself by thinking weirdly. And that's partly one of the things, because when I was little, I made these notes. They made sense to me. They made no sense to everyone else. But when you get older or when you as time goes, you, like, yeah that makes sense to me. So it's staying I mean its cheesy, but staying true to yourself and to what you what your vision is. So that's yeah. That's probably one of them. From the top of my head.

Harpreet Sahota: [00:47:44] I love that, that's absolutely an amazing way to put it. I think that's really the basis for creativity is taking two things that maybe on the surface of it don't look like they belong here or don't relate to each other, but then combining them in new ways to produce something completely different. That's kind of.

Camilla Pang: [00:48:02] Yeah, exactly. That's a great way of seeing it. And also, sorry, just want to not judge yourself for being obsessed about something. For example, if you obsess about this book or this or this link, go for it. Be obsessed about it, because this is how you get stuff done. And this is how you get to solutions. I think a lot people, especially when they're adults, they see obsession as something that is bad or chaotic or, I don't know, unreliable, I dont know what they think. I'm not an adult. And they try to silence themselves because they want to feel regulated. That's actually one of the important messages as well. So which is combined just to put that in.

Harpreet Sahota: [00:48:37] I like that. So let's jump into a quick lightning round here, starting with the first question. What is your Data science superpower?

Camilla Pang: [00:48:45] Right. Okay. I'm not very good with numbers, but I am very good at being able to simulate different models in my head simultaneously and know how they link back to algorithmic logic.

Harpreet Sahota: [00:48:55] That is one hell of a superpower. So if you could put up a billboard anywhere in the world, what would it say and why?

Camilla Pang: [00:49:02] I know I don't like that one. I'm not good with advertising.

Harpreet Sahota: [00:49:07] Yeah,

Camilla Pang: [00:49:08] Sure. Yeah. You know what? There it is. I'm not good at advertising 

Harpreet Sahota: [00:49:14] Thats perfect. I love it. So what's something you believe that other people think is crazy?

Camilla Pang: [00:49:20] A lot of people discourage reaction either from themselves or from other people. I don't know why I feel like this is due to the fact that we just we think we judge people based on this one reaction as opposed to seeing how a person evolves. So basically, it's a signal, but it's a positive signal. A negative signal. If I was really excited, she'd be like, oh, calm down, Millie, if I'm crying. Ah millie is being dramatic again. So I feel like a lot of people get scared of this intensity of reaction that we naturally hardness as humans because it's instinctive, which ironically is something that we're trying to get the computer to do. So that's the one thing I find really weird is that we're trying to suppress our instincts, to react, to look like we are logical.

Harpreet Sahota: [00:50:02] So what would you say is the most bizarre aspect or quality of human nature?

Camilla Pang: [00:50:08] That kind of. You know, we always seek conformity. I have no idea why, because, I mean, if we were to look at evolution. Cancer doesn't believe in, you know, conformity. If anything, it's a branched evolution. And this is something that I feel like humans naturally have that we naturally try and oppressed so that everything is coherent. That and making meetings about meetings. I've never understood.

Harpreet Sahota: [00:50:29] I like that. Meetings of our meetings. Yes. So another point about conformity. Do you think conformity is distinctly different from wanting to belong to the tribe or be a part of the tribe? Are those two kind of the same thing?

Camilla Pang: [00:50:43] I feel that we try and make conformity; its a critical mass effect than a bystander effect for a common cause. And that's fine. But when it comes to the point of exclusion, will you feel excluded? That's something else. If you feel like you don't fit or people like you need to be the shape, that's when it gets a bit sticky because you're like, well, I don't completely agree with that. And if you think about all these different categories of different boxes that humans live by not wanting, people are congruent. So to be able to be a conformist or completely in the middle of everyone, you probably don't exist. And if you did, you'd be on your own because, you know, it's no one's no one's normal.

Harpreet Sahota: [00:51:18] And you talk about this new book. So those who are interested to pick up the book, you go into crowds and individuality as well. So that's really interesting, really. So what is an academic topic outside of Data science that you think every Data scientist should spend some time researching?

Camilla Pang: [00:51:34] I'd say it depends on the Data scientists in question. For people like me who haven't started out as coders and are wanting to and hopefully learning a more about code, but also how Data science is. As an art and also what Data on its can do beyond what you think you can do. So I feel like it's a bit more the philosophy as well. If I were to get eggy about it. It's to read around the subject. And that isn't just like, oh, read another language. It's like no read philosophy. What are you to learn philosophy? Well, to be able to simulate psychology of different types of areas and different types of high order structures and how they localized down to the different people, then surely you're going to need to know the structures of hierarchies that exists. I see hierarchy. I mean, levels of abstraction, but also the interpretation. So actually, I think learning things that are unrelated, such as philosophy, psychology and art is something that is very fun to do. It's very inspiring. And it can also make you look at your work a lot differently.

Harpreet Sahota: [00:52:31] To quote Marcus Aurelius - What could guide us? Only philosophy. I 100 percent believe that. Philosophy is definitely one subject I am very deep into right now.

Camilla Pang: [00:52:41]  Yes great isn't it.

Harpreet Sahota: [00:52:42] It is. Yeah. So what's the number one book fiction, nonfiction or maybe even one from each that you would recommend our audience read. And what was your most impactful takeaway from it.

Camilla Pang: [00:52:53] Nonfiction book. Is it really hard? Because I actually love reading. And you learn something from every book. If I had to have one, I remember it's called Critical Mass By Philip Mass. It came out quite a few years ago, I think in 2004 or something like that. And I read it in my first year of uni in 2010. And I actually loved it. It gave me the confidence to be able to realize I can link things and I can link things. But that me linking things makes sense and is also desirable. Oh, so I can link science with psychology and psychology with physics and and then politics. Yes, you can. And so this book, I read it quite I mean, it's quite chunky thing, but it's definitely worth it. It discusses physics, politics and biology, graph theory and in such an accessible way. And also, agent based modeling is great. It's a really good book.

Harpreet Sahota: [00:53:44] I'll add that to the show notes and. 

Camilla Pang: [00:53:46] I'm not disagreeing with you was like, yes, it is.

Harpreet Sahota: [00:53:49] Pick up pretty much every book that my guests recommend to me. And I've got something like 40 unread books sitting on my shelf right now. But luckily, I tend to read a little bit faster, blast through them. But I'm looking forward to this. And something that's kind of on a related note to it, you just mentioned there's a couple of books. One is called Range. And one is called the self-made billionaire, in fact. And essentially the premise of those two books is that being able to take two unrelated ideas, put them together into something new, combining them in a new way, is what drives progress forward in premature every field.

Camilla Pang: [00:54:23] Definitely. Definitely. And I think that's where a lot of the new information evolution comes from is joining things unrelated for you to create something bigger. So, yeah, fiction. I can't answer that one. I'm afraid of fiction, but I've started reading it. But at the moment, I know I haven't read enough for me to feel like I am informed, but I actually love Normal People by Sally Rooney but that's my personal preference.

Harpreet Sahota: [00:54:47] So I'll definitely add those the show notes.

Camilla Pang: [00:54:51] It might not be your cup of tea, but you know.

Harpreet Sahota: [00:54:55] I don't have too much fiction in my bookshelf. I think the only fiction book I have is The Virtue of War, and that is a book written from the perspective of Alexander the Great through his conquest. So it's like a historical fiction. I guess in some sense. 

Camilla Pang: [00:55:09] Yeah

Harpreet Sahota: [00:55:09] So if we can get a magical telephone that allowed you to contact 18 year old Camilla, what would you tell her?

Camilla Pang: [00:55:17] Keep doing what you're doing. It'll come handy later on. People call you crazy now, but in our guess, because you don't fit in a system, that doesn't mean you weren't born to make a new one. And I feel to have that confidence is something that I wish I had when I was little. But once you can see that with every child. I was actually quite confident teenager in my own way. But it's just carry on memory. I don't regret doing anything other than I. I wouldn't be who I am now, you know, Bayesian. So just to reassure her that everything's gonna be OK basically.

Harpreet Sahota: [00:55:49] So what's the best advice you have ever received?

Camilla Pang: [00:55:52] I've got two bits that I feel like is that I can repeat myself on a daily basis, really. Nothing changes if nothing changes, which is really simple. But then you're like, oh yeah. Been doing the same thing. Wow. Why aren't I getting different results? Well, it's actually the definition of insanity. But when you're writing or communicating. Often people get writer's block, so to speak. And the very thing you are afraid to write is actually the very thing you should be writing because other people are going to be feeling it. But to have the bravery to communicate it is another thing that is to relate to things a lot people are scared of but don't have the guts to articulate. Now, that's great.

Harpreet Sahota: [00:56:30] Now I love that. Yeah. If you are feeling fear about something, that is a good indication that that is the direction that you should go towards. So what song do you have on repeat right now?

Camilla Pang: [00:56:40] Currently I really like Massive Attack and Moby, currently at the moment. I'd say Teardrops by Massive Attack in is quite retro, but always does the job.

Harpreet Sahota: [00:56:55] It's a good track. It's also the theme song for a TV show House M.D.  House M.D is about. He's a doctor. It's essentially he's a doctor. He's modeled after Sherlock Holmes.

Camilla Pang: [00:57:08]  Oh does it got Hugh Laurie in. 

Harpreet Sahota: [00:57:09] Yeah that guy. Yeah.

Camilla Pang: [00:57:11] Yeah. Oh, yeah.

Harpreet Sahota: [00:57:12] So what's next on the horizon for you? Any new projects? Any new books?

Camilla Pang: [00:57:17] Yes. So basically, I'm currently writing another book. But when I say that it might not turn into a book. It might turn into three books. I don't know yet. So I'm writing and assembling bits together. How I've always done. And from that decipher when they're going to go, when things crystallize more. I'm also speaking to TV production companies to see what we can. It is really exciting, actually. I mean, nothing's finalized yet. Nothing's approved yet. But it's great to have communication to see what can be possible to outreach. The message of not just your diversity, but also, I guess, a philosophy of science in everyday life. And just being able to get there.

Harpreet Sahota: [00:57:54] Definitely some exciting things, exciting things on the horizon for sure. And if people wanted to pick up your book, where could they find that?

Camilla Pang: [00:58:02] I guess you could find it online. I just tell people. Amazon or Waterstones. I mean. Yeah, and just Google it. Really? You find it.

Harpreet Sahota: [00:58:10] If people wanted to connect with you and find you online, where could they do that?

Camilla Pang: [00:58:15] Look at my Instagram. That's kind of where I kind of post stuff. Instagram. Twitter at LinkedIn. Yeah. Those are the three that I am mainly used to be honest. And you want to shout out your handles on those. Yeah. My Twitter is @millzymai. And my Instagram is Millie_Moonface. Dr. Camilla Pang on LinkedIn, I think.

Harpreet Sahota: [00:58:47] Yeah. So definitely add those to the show notes as well to guest profile. We've talked about. Sure. Social media profiles and stuff, but how could people actually connect with you. 

Camilla Pang: [00:58:57] By being real. Humor and curiosity go a long way, but the most important factor is for them to not be afraid to react. Or for me to react because. And also, indifference is quite literally a flat liner. React and be human.

Harpreet Sahota: [00:59:11] I love it. Dr. Pang, thank you so much for taking time with your schedule to come on the show to talk about yourself in your book. I really enjoyed it. I really appreciate you staying up late. I think it's quite late for you in England. So. 

Camilla Pang: [00:59:25] Thank you for having me on here. I really enjoyed it.