#196 Aran Khanna wanted a shorter feedback loop
⚠ The following transcript was automatically generated.
❤ Help us out, Submit a pull-request to correct potential mistakes
Aran Khanna 0:00
What was a useful skill in 2012? It's not necessarily a useful skill, or as highly in demand and 2021 a lot of software and solves come out for the biggest problems of the day. And the industry moves on to new problems. So I think that idea of constant reinvention and constantly finding within that new frontier, which is always moving as a developer, so specifically, what are the new things that are big problems, and that are at the intersection of pain plus your passion that you could go and work on next. And I think finding the next thing, be it within the context of your existing business or even jumping careers into a new industry or vertical. And that spirit of reinvention is really important to keep alive, not just the beginning of your career, but I think throughout your career.
Tim Bourguignon 0:41
Hello, and welcome to developer's journey to podcast, bringing you the making of stories of successful software developers to help you on your upcoming journey. I'm your host team building on this episode 196 I receive around Canada around is an engineer, an AI scientist and a troublemaker. After working for AWS. He embraced intrapreneurship. And he is now the co founder and CEO of Archie era, a company aimed at continuously and automatically managing and de risking your cloud resources around. Welcome.
Aran Khanna 1:18
Thank you so much for having me. It's great to be here.
Tim Bourguignon 1:21
It's my pleasure. It's my pleasure. But before we come to your story, I want to thank the terrific listeners who support the show every month, you are keeping the dev journey lights up. If you would like to join this fine crew and help me spend more time on finding phenomenal guests, then editing all your tracks, please go to our website, Dev journey dot info, and click on the Support me on Patreon button. Even the smallest contributions are giant steps toward a sustainable dev journey. journey. Thank you. And now back to today's guest. And as you know, the show exists to help the listeners understand what your story look like, and imagine how to shape their own future. So as always, as usual in the show. Let's go back to your beginnings. Where would you place the start of your
Aran Khanna 2:13
journey? Yeah, so I think my dev journey would have to start when I was in high school. So to back up a little bit, I was actually born and raised here where I'm calling in from in Seattle, Washington, which is and very much always has been throughout my life on the 90s and early 2000s, of Tech City with Amazon and Microsoft being here. And my parents actually worked both of those places. But funnily enough, in high school, I was a biology nut, I actually really wanted to be a doctor, I worked at a biotech company doing work on a lab that that was actually my first I wouldn't call it my first job ever, it was kind of an unpaid research internship. But that's what I was really interested in excited about. And funnily enough, while I was interning at this company that was using genetic engineering to create silent and bacteria that would produce biofuels, which was a very interesting project, I actually met someone who was doing computational biology work in the cafeteria. And he was working with Java to do genetic processing of our processing of genetic data and trying to draw inferences. And we would get to talking about our experiments. And I was saying, hey, go put this pipette and this tube and shake it up. And then two weeks later, I'd have results. And he's like, Oh, two seconds, I get results from my experiments. And then I run them again. And then again, and again. And I'm done. At the end of the day, in terms of producing enough data and research to put together a paper or some series conclusions while you're there, twiddling your thumbs waiting for your little algae to grow. And at that point, they're impressionable high schooler I'm like, What the hell am I doing with my life, if I was going to try and make an impact in this world, I would need to have a velocity with which I could build and discover, but it's much higher than like you get sitting in a lab bench. And that's really what sparked my interest in computer science. And while I took a few classes online, actually to teach myself Java and learn a little bit more, I really didn't get deep into it until I started college. And that's when I actually started working as an intern. And that was my first job ever over at Microsoft. And this little project in the research team that eventually became Windows Azure. So really saw kind of the birth of the cloud within Microsoft working on some of the fundamental systems there. And I think that's what really catapulted me into a focus in my career on technology, and not just kind of enterprise cloud technology, but really where my passion was, and still is on understanding kind of these big platforms that have become central to daily life, things like Facebook, things like AWS and Azure, and helping customers sort of navigate and build on top of those things. So in my kind of research life, I actually did a lot of work in the privacy space when I was in college, working on exposing, if you will, some of the issues with how platforms like face treat and expose data about us on that actually got quite a bit of press, but really my journey kind of ping pong all over the place even within the dev world. I went from privacy research to working on machine learning, with a professor at Carnegie Mellon joining his startup after school that, again, got acquired by a big cloud provider by AWS and catapulted me into working on the early teams there that built out a lot of the AI services that are a big part of the platform today stage made deep lense, which are actually pitched and launched at the company, a lot of the managed machine learning API's, like recognition, their computer vision API to find objects and images and their speech recognition API as well, that's used to power things like Alexa. So really worked all over the place there. And I actually started as a researcher, I didn't actually start as an engineer who was building product, because I wanted to learn more about what is this deep learning thing? What are the limits? How efficient can we make these networks, which is really my area of research. But being in a big company like that, and with all the kind of pull of the management, you naturally get forced into building product with some of that research work. So it's very interesting. And I think seeing a lot of the customer problems from both sides of the user of cloud platforms, then as a builder within them, it drove me to create our chair, really start my entrepreneurial journey. So really not a straight line. And all I had tons of different interests got pulled in tons of different direction, but incredibly happy that I landed here. And I can build stuff within a day that if I was even in synthetic biology today, it would take me weeks to do I think I made the right career choice,
Tim Bourguignon 6:19
at least from that standpoint. Exactly. I'd like to come back to us very well, you point. You said, Okay, you started learning on your own maybe Java, etc. And then boom, internship at Microsoft in on the what became Azure. How did that happen? Oh, yeah, from knowing nothing to being a microphone or Microsoft?
Aran Khanna 6:37
Yeah. So my first year in college, I actually took a number of computer science classes. And I was super passionate about building things, right. So I had a ton of side projects that I actually worked on. And I think that was a lot of the reason why, even given my experience in this space, that specific team that I applied to, and was interested in, took me on because they had a track record of building things that were useful and interesting. And one of my side projects at the time, and I wish I continued building this was back in 2012, we all had these phones that were able to emit and pick up incredibly high pitch like ultra frequency noises. And we created a little app actually during a hackathon that would let you encode data over songs like a QR code almost but audible that a phone could pick up if you'd say play it over loudspeakers. So there was things lab that I think were interesting. And as I started to work on those and put those out on GitHub gives me enough of a track record where I could actually go and work at a more established corporate and get some of that formal training. Because I would say even at college, where I was almost like the European system, where you're at where you were at, it was very academic, right at Harvard, computer science is not a practical thing like it is down the river at MIT. It's about the theory. It's about the mathematics. And really even that was my focus there. I studied more the algorithms and the mathematics than the actual implementation science, which I had to go to industry to get,
Tim Bourguignon 7:56
what was the path you envision for yourself at that point? For me,
Aran Khanna 8:00
I honestly had no no path in mind, I wanted to find things that were interesting and explore. And basically, at the beginning of school thought, whatever is interesting, and I'm passionate about and whatever it takes me for my internships, that's where I'll end up going. So be it on the big tech side, be it on startup side, which I also explored, I was kind of at and just got pulled in a bunch of different directions. And whichever one kind of rose to the top of my stack rank of the moment became my top choice. And I think generally, as I went through school, and what I would recommend to other folks going through that same sort of journey, trying to figure out what to do, it is very much an explore, exploit sort of problem and the front, at least half of your schooling experience should be an incredible amount of explore. Because there's so many interesting topics out there, even within a field like computer science and engineering, which 2012 13 There were so many fields, you could still go into with that degree, you could be a quant in finance, you can be in computational biology at the Broad Institute, you could be in tech, or startups and entrepreneurship. So even within one chosen vertical, there's so many paths to explore. I think it's incredibly important to go and do that.
Tim Bourguignon 9:08
Absolutely. Did you decide then to pursue this exploring in a PhD in research? Is that what you mean when you made until you were a researcher or?
Aran Khanna 9:17
Yeah, so actually, after school, I'd done my last internship with which is after actually got fired from Facebook, which is another story, privacy research, which is another fun thing I explored before getting into machine learning research. But it was with a professor from Carnegie Mellon, who is fairly well known. And essentially, as I was coming out of school, his pitch to me was come work at my startup. And we'll pay you to get a PhD, essentially, as an engineer, you can write papers, you can build open source software, which was what I was doing at that startup before it got acquired by AWS working on the MX net product, which is a deep learning open source Apache framework, and you can have the best of both worlds. So that was really appealing to me. I wanted to dive more into industry, particularly on the small team kind of startup side of things where you could build a product From scratch really kind of from the ground. And I also wanted to understand more about deep learning and machine learning because it really lay at that intersection of computer science and mathematics that I had really gotten into in, in college from an academic perspective, but not from an implementation perspective, particularly because the resources required were so massive to go and train these models. And obviously, the costs have come down quite a bit, but it's still expensive to train deep learning model.
Tim Bourguignon 10:25
Yes, yes, yes. Okay. So that was basically alongside your your take your professional take journey from
Aran Khanna 10:30
it? Yeah, well, very much a part of it. Right. It was kind of a great opportunity that I'd found in the intersection of academics and an actual implementation,
Tim Bourguignon 10:38
we have to come back to what you just said, in one sentence, and then discarding it right after you were fired from Facebook.
Aran Khanna 10:45
Oh, yes. So I'm really excited about what happened in 2021. Because I feel like I was ahead of the curve in 2015. But part of my early tech journey before I got into machine learning was actually in privacy research. So one of my professors at Harvard was actually the former CTO over at the FTC, which is the Federal Trade Commission in America that regulates companies like Facebook, and has handed down Big fines before for things like data protection issues that they've had around their customers been working with her, I'd actually started to notice a lot of things in the apps that I was using with my friends at Harvard, particularly apps like Venmo, and Facebook Messenger, were kind of the defaults for your sharing settings, were always set in this way that created a very pernicious and invasive data leakage problem across the board. And people had written about this before in some respects, but my thought process as a developer was, look, people don't really understand that if in this case, it was like messenger, if they leave the default on every message they sent to any group chat, no matter how many people were on it, or who was on it would actually attach their location, their immediate location by default. And what they didn't actually realize, because some people like yeah, whatever, it's only one message. But what they didn't realize is, over time, if you actually added up all of that data and plotted it on a map, you could come up with some incredibly insidious insights about that person, their schedule, who they're communicating with, and hanging out with. And so what I did was I actually built a Chrome extension that sits on your browser, pulls in all of that data from you and your friends, and then shows you exactly what that default being is actually causing you and your friends to expose about themselves to really the world, you don't have to be friends with someone to go and get this data from a group chat you're in with them. And this is something that can be done with a pencil and paper, I just did it with code. Easy, right? And actually, it was kind of funny, because I thought, this is going to be a service to the users of Facebook and Facebook would be happy that I was helping educate and show people what they're doing with their data and how they can if they think this is a problem, turn it off. And I actually had an internship lined up that summer. So I thought Shit, they might offer me a full time job because of this, because it's what's going to have impact and all those things that you want, as a developer at scale on these features that are important to us. And so really what happened was I released this long set of paper and blog posts and a GitHub project, completely open source, right. And they had called me and asked me to take it down. And I'm like, okay, sure I did, I've made my point. And the day before I was supposed to start my internship, I had already signed a lease for three months for $10,000 a month in San Francisco, and I was banking. On joining that Monday, they called up and said, Hey, by the way, you didn't act in the best interest of Facebook. So we're letting you go. And what was crazy to me, and I think what is now finally being more publicly talked about the company is the fact that they clearly do not think that the people who use their product are the customer, and they don't treat them like the customer, they treat them like freakin cattle. So the fact that I was educating people who are using the product and how to use it more safely was a negative to them, because it caused a PR issue. Instead of them seeing it as Oh, you're actually benefiting our customer, the customer to them is the bng paying for as though it's really kind of an issue with the culture. Actually, I've talked to people at Google were very high up after this happened. And they would say, look, the reason that Facebook's in so much more hot water than Google, despite us potentially having even more data is it's a cultural issue. And I think I saw that here. The cultural issue there is you do it in Facebook's best interests or you're out and the issue at Google and people make fun of it. And yes, there's legitimate criticism, but fundamentally don't be evil is something that they take to heart and apply to the users of their products. So I think Facebook is very much a different beast. And that experience to me, in a way proves that without fundamental change at the top and culturally. They're going to continue having these issues, be it on privacy or on misinformation, teen girls and depression, what have you. This is not going to be an easy solve until you solve the cultural problems there.
Tim Bourguignon 14:49
If it's solvable at all, I mean, you go I'm glad to quit Facebook 10 years ago, I drink during college, it was something that I dropped To me, that wasn't a good thing, although I lost a few friends in the process, but that's the way it is. Okay, so no internship for you, I Facebook, yes. But in hindsight might be good,
Aran Khanna 15:11
might have been a good thing in hindsight. And actually, that wasn't the first time I did that I continued my privacy restripe and do privacy advocacy to this day alongside things like machine learning safety, kind of outside of my nine to five with our chair. But another project I did was actually with regards to Venmo, which is the peer to peer payment app that also had these fairly invasive defaults where every transaction was public. So I built another app that went and scraped in all of that data, and actually made a map of your payments and the volumes and you could pretty easily see things like, Alright, these are all the people who are hanging out together. These are the people who are going on a first date. These are the people who all are a member of this club. So this is not a purely a Facebook problem. I think Facebook is probably the worst offender and the one that is at the most scale is a fairly common issue that I was seeing back in 2015. And now it seems paltry in comparison to other issues we're seeing now with recommender systems gone awry. And misinformation, and obviously the slot machine design on apps like Instagram. So I think this is a very interesting area and one that while I haven't found a great way to monetize and build a business there is important that we continue to talk about and push on, because otherwise, there's really no checks and balances for companies that have kind of the power over the content on the end of the device.
Tim Bourguignon 16:28
Right? Amen to that. Is it important for you, when you pick up a side project like this, to really have a practical end goal in mind already knowing where you're going to apply this, what it might bring, etc.
Aran Khanna 16:42
So actually, I, the way that I think about it is it's like 25% play, and then that play helps you find the other kind of focus area for the 75% of the time to go and do an implementation with working backwards from a customer facing or user face again, goals for the part of Marauders Map, which was the Harry Potter inspired extension that scraped in all of the data on Facebook Messenger displayed where your friends were in, in the castle, so to speak, a lot of what I was doing kind of upfront there was just looking at is the state of compelling doing the data exploration pulling it in and kind of playing around with it. And then once I understood that it could be compelling on a few use cases, working backwards to build the actual end product that would be compelling to a lay man. Now someone who's not a Jupyter Notebooks expert who could go and slice and dice and find the histograms. Right. So I think that's kind of a common pattern that not just my side projects have followed, but even building our Chehra follow that same sort of pattern, what I thought early on was, hey, as I launched new services at AWS, and saw our biggest customers grapple with trying to manage these services, particularly with the complexity of the costs, I saw this massive misalignment between the cloud vendors and the platforms and the 64 gigabyte pricing files, they would send to folks to try and parse through an Excel and a customer who just wanted to get a good deal. And they were used to going to Dell and HP and negotiating the price of the box. And now people are just dumping a terabyte of data on them and say go solve this combinatorial problem. So it was things like that. And I think it's kind of common even to draw the line between what what Facebook was doing trying to pull the wool over the eyes of its users with respect to privacy and sharing defaults. And really what a lot of these cloud vendors are doing in terms of trying to the wool over the eyes of their users, with the complexities of their pricing and the management of pricing contracts within their system, they set the rules on how to bill you, it's your job to navigate them. So when I saw that and picked three or four areas where I thought there could be impact, it was a lot of play and experimentation with kind of small early customers for we found out alright, what was that actual automated workflow and and that we need to work backwards from that can really alleviate the solution in a 10x better way than the manual processes and the back and forth and spreadsheets, which is really still the norm in the industry today.
Tim Bourguignon 18:56
Okay, how can I picture your product right now is really analyzing what I'm using coming up with with optimization is telling me what I should be using given the goals I have in mind. Because,
Aran Khanna 19:07
yeah, it's kind of both. So the way that I think about it is there's this old world that I constantly saw in companies using Amazon that was trying to sell edge maker and these other services, we were building on the ML side to where they had individual engineering teams at some centralized Cloud Manager or financial management. And they would constantly be going back and forth over the course of months. So the engineers would go and label and consolidate their infrastructure, the finance team or the centralized management team would report on it, then the engineering team would have to go and terminate stuff that was unused. And then the finance team would go and analyze workloads going forward based on business constraints, and then the right sizing would happen and then the forecasting would happen, and then the validation of the forecasts would happen. And then finally, they would go and purchase commitments, which takes a whole bunch of parsing and might be wrapped up with the top level commitment that they have with Amazon or Microsoft, and then they would have to go and manage those in a real time fashion. And so it's just incredibly high friction back and forth. And all the tools that I was even selling like third party vendors and consultants would just try and streamline pieces of this process, what we're really trying to do is change the process altogether, where you can have one person who sits there, and does a fully end to end automated kind of tagging, forecasting and right sizing assignment, and then actually purchasing commitments, and uniquely what we're able to do because of automating a lot of this workflow is sell insurance against those commitment purchases. So if you don't use them, if things change, we'll actually buy that back from you. So we try and not just change the process to one that can be streamlined and automated through software, but one where that streamlining and automation gives us so much confidence that we can actually share risk with you.
Tim Bourguignon 20:49
Stay with us, we'll be right back.
Tim Bourguignon 20:52
Hello imposters, if you work in tech want to work in tech or are tech curious in any way you'll want to listen to this. We've launched a community of professionals who come together to share information and advice about jobs, roles, careers, and the journeys we all take throughout our lives as the designers, builders, fixers investigators, explainers and protectors of the world's technology. We call it the impostor syndrome network. And all are welcome. So find the impostor syndrome network podcast wherever you listen to find podcasts, and look for the isn community on your favorite social platform. Hashtag impostor network.
Tim Bourguignon 21:34
Who, okay, I want to dig in there. How can you share this risk? You don't have any upside, you cannot buy a bag? Because it's well,
Aran Khanna 21:41
actually, you don't? Well, you can. But funnily enough, you can, there's actually an API, if you search, for example, the AWS standard ri marketplace, you can see that there's actually a peer to peer exchange mechanism for this. And in fact, what's very interesting is that is things like capital expenditures in the old world, you would need that box to be resellable. So a lot of companies actually make the argument that if you can just resell a reserved commitment to a box of Amazon, you should be able to treat it the same way from a capitalization perspective. So some interesting thing around why this must exist and why it's beneficial. But the net of it is that there's very little liquidity in that environment right now, no one can be guaranteed if they list something there, that people are going to buy it because, frankly, the specific region and the specific instance type and size at a specific time, and it's all very specific, and there might not be demand for that. But because what we do is we help folks not only plan for when capacity will be available, but also automate their purchasing cadence, we have a much better sense of when the future demand is and we can underwrite that risk. And anything that we have to carry in the internet, we charge premiums to our customers essentially the same way any insurance provider would that are much lower than the incremental savings of, of buying these long term commits through us. So that it is the system works out such that our customers save more money, they have more flexibility. And by having that visibility to make the market between them, we're actually able to make a little bit of a profit, but really service our customers in a much more robust way and take more risk, because they're really sharing it amongst each other. We're not the ones just loading it all on our balance sheet.
Tim Bourguignon 23:16
It's fascinating, I discover a new world I didn't know existed, this is this the core of your business, or this other side of optimizing how you work with your boss. Yeah, just to finish my thought I see it like like you described Facebook at the beginning. It's really two completely different personas that you have in mind if you tackle one or the other. So how do you do that?
Aran Khanna 23:42
Yeah, I mean, it has two completely different personas. And this is why I think the core of our business, frankly, is because the market is so immature in terms of people just figuring out how to manage this, how to navigate the price of spreadsheets that are being sent their way by the vendors, when they're making big purchases and commitment that the first bit of our business, we're just saying, Hey, here's the process to manage this, here's a new problem that you're dealing with, here's the new process to manage it that needs to happen for us to go and sell the insurance on top of it. And that really ends up being the core of our business actually going in and educating customers educating developers saying, hey, going and clicking the button and buying the lowest savings rate, highest flexibility, commitment might get financed to stop breathing down your neck, that it's too expensive. But it's not actually the long term choice for your business. And then getting the finance team to say, hey, if you're actually operating the cloud, the same way you were doing on premises, you should be building projections, you should be using data to figure out what is future usage and spend, and actually building budgets in for that, instead of leaving it as some high variance black box. So it really educating customers on what they should be doing. And then empowering them with the tool to do it becomes the core of our business just given where the market is today. And then what's really great is the customers who are sophisticated and don't need that they can take advantage of the insurance and the customers who really do need that. Once they're ramped up, say hey, this is a really great additional benefit. I can actually Go and be more aggressive and execute on more interesting strategies because of this additional optionality, I get running my process through this system, and what
Tim Bourguignon 25:08
is the archetype if there is one of the customers you're shooting for?
Aran Khanna 25:13
Yeah, so the ideal folks for us actually end up being people who are put in particularly to solve this cross functional challenge, there's a new world called fin ops that you might have heard of, that are people who sit between engineering and finance teams and own this problem. So they end up being the best sort of champion or user with a customer for a solution like ours, but that's actually very few and far between. In the spectrum of, of different sort of organization, not a lot of customers actually have that role scoped out. So we often work with site reliability engineers and DevOps engineers on one side. And then we work with DNA finance, procurement, as well as potentially CIOs or VPs of finance on the other side to really reconcile and build this process didn't ovo and that organization and kind of create a fin ops sensibility and process an organization where one just didn't exist before? And
Tim Bourguignon 26:03
what is the size of the company that would be attracted by those products? I mean, if you're too small, I guess I guess developers are all all the engineers are managing all this on their own, there is no discussion there. And if you're becoming really heavy, then you probably have so many levels in between that it becomes harder to give that out to the outside. I know. What do you say?
Aran Khanna 26:24
Well, actually, I would say on the low end, you're right, and people who are just getting started, you might just have the developers handled this, it's not a big line item cost, it doesn't really matter much. Now that could change. If you're doing massive machine learning workloads, you might want some governance around it, you might want some predictability and insurance even against those massive spiky spends from things like GPUs. But by and large, usually, we stay away from those folks on the high end. Actually, it's folks like the Netflix's the Spotify is the Airbnb is the world that have so much margin. And so many developers were there, just go and build the capacity planning team, they're gonna build it all in house, they don't want to buy software or come up with solutions that are not in their sort of core stack that they're going to build, maintain and push out. So Netflix has a number of open source solutions that they even maintain today to manage stuff like that. And I think they're also just massive, right? So we can't even take that much of their spend and insurance just from a volume perspective. But it's really folks in the middle. And we have customers ranging from series D startups where finance is just coming in. And they're really starting to care about margin all the way through to companies that are in the Fortune 500, like fortive, which is a big industrial company out here in Seattle, that has a number of different subsidiaries that run on the cloud underneath them. And they're trying to put in some sort of centralized fin ops management so so really on both ends of the spectrum, and even in the middle, folks, were just IPL. FreshWorks and garden health are also sort of in the in that spectrum of customers that we serve. So really, it ranges but folks who are very small and very large tend to be outside of the scope of would be a great fit for us. And it's really people who have that, that start of the dialogue between finance and engineering, the early stages of building it out or just have one or two fit Ops had starting to manage this problem that are really great fits to put this process in place and take advantage of a lot of the benefits that we offer. Okay,
Tim Bourguignon 28:10
if I put it in my words, that's basically the phase where you're slowing down throwing money out the window, saying, well, it has to work, let's just go in and make it worth it sounds like Okay, now we pretty much know what we need. So let's try to make a lien. And that's where exactly this concern comes in. How much are we actually paying for this thing? paying too much. And that's where finance will probably become way more interested in what's happening. And so,
Aran Khanna 28:33
and even questions like, Hey, should we start writing off some of our r&d spend in the cloud, things like that finance really cares about and understands. And developers, they don't know that this is actually something you could even do or what's our cost of capital is how much should we be spending upfront? These are things that engineers are understood it and write the python script, the calculator, they would, they would totally get it. But this is not something that's in their wheelhouse. So it's really this cross functional area. It's quite quite a white space right now, where we're trying to come in educate folks and really that process support.
Tim Bourguignon 29:05
How was your role evolved in the past year? Or maybe? Do you still come to to coding at all? Are you the machining Chief, are you as a visionary? How do you describe all this?
Aran Khanna 29:18
Yeah, so I mean, actually, I think Steven today I'm number two in terms of commits in the code base. That being said early days that a lot of Cody had built the first few prototypes and brought it to customers alongside my CTO who's still codes and is absolutely a wizard, so I would not want to take that away from him. But that being said, over the last year, I've definitely really fallen off a cliff in terms of the amount of code I push it and really ramped up in terms of product roadmap reviews and setting kind of one month, three month, six month, sort of goals for the team doing maybe one or two PR reviews per quarter. But you know, largely working on hiring and getting managers in place and having them put in place mentorship and review processes and endurance for the team and then a lot I have sitting in front of customers and reporters and podcasters and talking about what we're doing and build and acting as a bit of a mouthpiece for the company as we grow. So it's really high variance. I do a lot of different things now. But the one thing that I don't do a lot of anymore, and it's kind of sad to say, is writing code. In fact, the most code I write is actually my off time when I'm just playing around with stuff for writing some trading algorithm for bitcoins. I don't know, just for fun to blow off steam though. Yeah, in my main job, I actually I don't think they would even want me to touch the codebase. Now.
Tim Bourguignon 30:31
I know that feeling to blow off steam and coming back to you literally blew off steam with Bitcoin. What are the kinds of pet projects that you have nowadays that you're scratching? And besides? Because you're mentioning, do you have any other No, I
Aran Khanna 30:45
mean, I do actually have a script that runs but I just maintain it. Now. I think some of the interesting stuff that I've been working on is got a whole bunch of IoT devices for for my myself basically, for the holidays. And I've been playing around with those and just seeing what the latest and greatest is on that side. And some of the different technologies that have come out from the Bluetooth side and the Wi Fi six side just trying to play around. Because I used to be into a lot of those networking technologies as well. What else am I working on? I think I've done a little bit of research. And I'm trying to set aside some time over the holidays to go play around with this idea of Dows. I think they're very interesting. And there's actually a lot of kind of interesting reading and things that you could kind of build around that. So I'm talking with some friends, maybe kicking the tires and doing something related to that in the next few weeks just to see how it worked. But yeah, I just like learning stuff and exploring new areas, like I said earlier, pretty ad and you could probably tell that by all the different stuff that I've worked on in the past. So it's fun, always energized and interested in stuff at work, because I'm always working on new things. But I think on the technical side, I also need to feel engaged and see what's out there. What's interesting. And what's awesome is we have such a big variety of customers that a lot of these ideas have like stuff to work on actually have like a BIG IoT customer, and I was talking to them about what they do. And like, oh, cool, like, there's so much new technology, I'm gonna go and kick the tires on this. And we have a crypto customer actually, and I was talking to them, and they're telling me about all these innovations. So early in the course of my job, I find a lot of cool things to go work on. Because our customers are always working on interesting things. And they're so diverse. And yeah, it's a great part of the job.
Tim Bourguignon 32:12
It looks like in the you cannot see it on the on the tape, but you have a big smile on your face. And while you were describing this, your eyes were lightening up. So it seems to be the right place for you. So what would you say the one advice that maybe you heard along your journey or that you you learned along the journey, and you would definitely want to give out again for especially for newcomers in the industry trying to find their way.
Aran Khanna 32:34
Yeah, so I think my my major piece of advice is going to be coming back to what I was talking about earlier, which is exploring, being able to take time to see what's out there. And what's interesting to you what you could be passionate about, because the world is changing so fast. And what I'm realizing even being in the cloud for five, six years is what was a useful skill in 2012. It's not necessarily a useful skill for as highly in demand and 2021, a lot of software solves come out for the biggest problems of the day. And the industry moves on to new problems. So I think that idea of constant reinvention and constantly finding within that new frontier, which is always moving as a developers specifically, what are the new things that are big problems, and that are at the intersection of pain plus your passion that you could go and work on next. And I think finding the next thing, be it within the context of your existing business or even jumping careers into a new industry of vertical that spirit of reinvention is really important to keep alive, not just the beginning of your career. But I think throughout your career,
Tim Bourguignon 33:32
it is very much indeed. Thank you for highlighting that again. So room, where would be the best place to continue a discussion with yours or start a discussion with you.
Aran Khanna 33:41
Yeah, so you can find me very easily at the archera website, a r c h e r a.ai. So if you're interested in learning more about kind of the intricacies of cloud pricing and cloud optimization and financial engineering and the cloud, you can find us there by booking a demo or scheduling a chat with myself or someone on my team. Or you can find me on Twitter at Ron Khanna
Tim Bourguignon 34:03
and we'll add both links in the show notes. So just scroll down and click and you'll get there around. It's been fantastic. I discovered a whole new field I didn't know existed. That is that doesn't happen so often. I'm stoked. That's really cool. Thank you.
Aran Khanna 34:19
Let me know if you have any questions on it. There's a lot of weird intricacies and interesting stuff in there. So kind of a black hole, honestly, but in the best possible way.
Tim Bourguignon 34:27
I'm gonna go and read a few things. First, I discovered this on my own and I'll definitely take you up another over.
Aran Khanna 34:32
Thank you very much. Well, it's wonderful chatting with you. And thanks so much for the time. It was
Tim Bourguignon 34:37
and this has been another episode of Devil's journey. We see each other next week. Bye. Thanks a lot for tuning in. I hope you have enjoyed this week's episode. If you liked the show, please share rate and review. It helps more listeners discover those stories. You can find the links to all the platforms forms to show appeared on our website, Dev journey dot info slash subscribe. Creating the show every week takes a lot of time, energy and of course money. Would you please help me continue bringing out those inspiring stories every week by pledging a small monthly donation, you'll find our patreon link at Dev journey dot info slash donate. And finally, don't hesitate to reach out and tell me how this week story is shaping your future. You can find me on Twitter at @timothep ti m o t h e p orca email info at Dev journey dot info talk to you soon