Working at Instacart during Start of Covid was Insane - Cofounder Story - EP8

Show notes

As we set to launch our product to the public, we take a look (or a listen) at cofounder Anthony Bobsin's journey that led him to this crazy startup, Dreamwell.

If you're new here, Dreamwell is an AI-powered software that helps companies find influencers who will actually generate revenue for them, and not waste their ad dollars.

Listen to Bobsin's cool adventure in how he got to where he is today, from his extremely unique and interesting time at Instacart during covid's first days (remember, everyone began ordering groceries due to lockdowns), to his first days at AdParlor...a full time gig WHILE at school full time!!!

Enjoy episode 8, friends!!

Where to find us

Anthony Bobsin: https://www.linkedin.com/in/anthony-bobsin
Kazzy Khazaal: https://www.linkedin.com/in/kazzyofficial
Xinchi Qi: https://www.linkedin.com/in/xinchi
Dreamwell AI: https://www.linkedin.com/company/dreamwell

Episode timestamps

Introduction [00:00:01] The speakers introduce themselves and the podcast, explaining that they are building a creator influencer marketing platform and documenting their journey.

Product Development and Closed Beta [00:01:17] The team discusses the progress of their product development and their plans to shift into a closed beta phase, where they will slowly roll out access to their product for certain people.

Coding Techniques and Focus [00:03:13] The speakers discuss their coding techniques and how they stay focused, including listening to music, using focus tones, and the concept of rubber duck debugging.

Gamifying Writing [00:07:22] Kazzy talks about how he gamifies her writing experience by communicating with different characters in her head to stay focused.

Suppressing Inner Voice [00:08:08] Kazzy and Xin discuss their different approaches to staying focused, with Kazzy trying to suppress her inner voice and Xin trying to stay present.

Bobsin’s Origin Story [00:10:44] Bobsin talks about how he got into coding, starting with hacking video games and listening to his dad's side projects, before taking courses in university and building his own projects to land his first job.

Computer Science Degree [00:15:17] Discussion on the usefulness of a computer science degree and the need to learn on your own.

Juggling Work and School [00:16:03] Bobsin talks about juggling full-time work and full-time studies and the challenges he faced.

Instacart [00:20:54] Bobsin talks about how he got referred to Instacart and the growth of the company in Toronto.

YC Company Growth [00:22:38] Discussion of the growth of a YC company, Instacart, and the adjustment to working with a larger team.

Instacart during COVID [00:25:11] The experience of working at Instacart during the COVID pandemic and the challenges of scaling up quickly.

Transition to Startup [00:28:47] Reasons for transitioning from a large company like Instacart to a startup, and the desire for more ownership and flexibility.

Remote vs In-person for Junior Developers [00:32:27] Discussion on the difference between remote and in-person work environments for junior developers and the challenges they face.

Hybrid Work Model [00:35:24] Exploration of the benefits of a hybrid work model, with a physical office for co-locating and a remote option for those who prefer it.

Company Cottage [00:37:07] Humorous discussion about the possibility of having a company cottage, such as a beach house or a house beside a lake, for the team to work from.

Product Progress [00:38:26] The speakers discuss the growth of their company and the experiences of working at Bobsin’s Time and Instacart.

Next Week's Guest [00:39:17] The speakers announced that they will have a cool guest next week and encourage listeners to follow the podcast on Spotify and Apple Podcasts.

‍

Transcript

Kazzy (00:00:01) - Welcome back, ladies and gentlemen, to the Startup Journey podcast. It's your boy Kazzy. We've got Xin and we've got none other than Bobsin. What's up boys? How you doing?

Bobsin (00:00:11) - Hey everyone.

Xin (00:00:11) - Hey guys.

Bobsin (00:00:12) - Doing well.

Kazzy (00:00:14) - So, um, guys, look, let's just, okay, so if you're new to the podcast, we're three guys trying to take over the world with software and get people's data. First party data. I'm kidding, I'm joking. We're building a creator influencer marketing platform. We're trying to help companies stop wasting ad dollars on influencers who don't convert revenue. And we're documenting our journey. Since we're documenting our journey, guys, look, let's just talk like openly, casually about where we're at right now, where the product is sort of at, we don't have to get into too much details, but like where, what, two months in essentially. People are listening to this right now that are probably wondering like, what does it feel like to have a startup? What does it take or where are you two months into it? I think people are gonna find our podcast just by the name of it, right? Mm-hmm. . So now this is the eighth episode I think, or something like that. So like, where are we at? Let's talk about it from a product standpoint. Product team. How are you guys doing?

Bobsin (00:01:17) - Yeah, so we have the name Dreamwell now...

Bobsin (00:01:24) - Jokes aside, we've been busy building the software product over the past few months and at the same time starting to engage with our target customers so that we can have conversations with them, get some feedback so that we can incorporate that into the product we're building. And now we're at the state where we're shifting into we call a closed beta, where we're going to open up access to our product for certain people that we permit. So the idea is we'll do a slow rollout to people that we're talking to and slowly get the product in the hands of these, these potential customers.

Kazzy (00:01:58) - And Xin, how are you feeling? How, how are you feeling when you're coding? How are your sessions? That's what I wanna know. I wanna know about your sessions. Like when you guys code, do you turn off your notifications? Do you say, don't talk to me for three hours? Do you like put a sign on the door like, cuz I know you gotta get into a zone, so let's kill some stereotypes and create new ones. So how do you code? Like what's your session like, Xin?

Xin (00:02:30) - It's like, um, how can I best describe my sessions? My sessions are like unending tasks working in a pressure cooker. That's, that's what my sessions are like.

Bobsin (00:02:44) - Sounds stressful.

Xin (00:02:46) - I mean, coding is stressful sometimes, especially when you don't see the progress of what you're doing, which is typical for backend stuff.

Bobsin (00:02:58) - Yeah. That's what happens when we don't get much user feedback yet, from a live product at least.

Kazzy (00:03:05) - Wow. Especially back end. You're right. Cuz at least on the front end you could see it.

Bobsin (00:03:08) - Right.

Kazzy (00:03:10) - Nice. And what about you Bobsin?

Bobsin (00:03:13) - Yeah, I work pretty well in a number of different environments. Um, but if I really have to build out like a, a feature that's, that's fairly lengthy, um, - do you guys know two friends, like the big booty mix playlists? So they're these two DJs that make really good mixes on YouTube. Um, and they're basically just like a mashup of all the top a hundred songs at a time. So like, they're so catchy and they're all blending into each other. Um, and that just like gets me going and that can be like really productive when I'm listening to those playlists. So I reserve those for when I really need to build something and push it out fast.

Xin (00:03:54) - So the music helps you kick into, uh, into mood. Yeah,

Bobsin (00:03:57) - I'm always listening to music whenever I'm writing code.

Kazzy (00:04:00) - Nice. Interesting. What about you Xin? Do you listen to music?

Xin (00:04:04) - I, um, no. It's funny cuz I listen to music when what I'm doing, stuff I'm familiar to. But if I actually need to get in the zone, I have to be like, it needs to be no sound.

Kazzy (00:04:19) - interesting. When I'm working I can't have music. I need, do you know iso chronic tones? Do you guys know what those are?

Bobsin (00:04:29) - No.

Kazzy (00:04:31) - You haven't seen these videos? It's called focus tones. Okay. You gotta have to listen to it. After this. I'm gonna send it to you. You're gonna think it's nuts. It's like frequencies that play, ah, it's like din boom, boom, boom, boom. Just like that. That's it. It goes on for three hours and

Bobsin (00:04:46) - I can't do those.

Kazzy (00:04:47) - It's nuts. Like I play it and I get zoned in and when when I mute it, that's when I know it's working. Cuz when you mute it, you're like, oh, I don't like this. Like I, my brain needs constant stimuli. So when I play it, it's like stimulating my brain. It's, it's, there's two sides, like on the headphones, one does one frequency, one does the other, and it's brain waves. It's like mm-hmm. proven to do stuff. So when I play it, I'm like, Hey, I know for the next three hours I'm not gonna hear another sound from anywhere. No car, no notification, nothing. I know that I'm free to listen to this noise for three straight hours.

Bobsin (00:05:21) - You know, there's a lot of value in that where you basically trained your brain to, to really focus when you hear these sounds. I had this specific classical composer in university called Dexter Britain. And I would only listen to his music while studying. And like, I, I still check it out once in a while and I get like goosebumps cause it's just, wow. It's so associated with my whole university experience now.

Kazzy (00:05:43) - Amazing.

Xin (00:05:45) - All the PTSD kicking back, eh?

Xin (00:05:48) - Well, I have a question. Do you guys hear like your own voice when you're, when you, when you're focusing like your thoughts?

Bobsin (00:05:54) - I had a, yeah, I don't know. Can you share what's the alternative Xin?

Xin (00:06:03) - You're speaking to not hear my own voice. Well, I don't know. It's like when I'm really in deep in the zone, I feel like I'm three person working. You know what I mean? Like I hear, I hear my voice like speaking to me, my thoughts.

Bobsin (00:06:16) - Yeah.

Xin (00:06:17) - Like this is when I talk.

Bobsin (00:06:18) - Yeah. Yeah. There's this concept in programming called rubber duck debugging where it's a concept where people would have this rubber duck, on their desk and then they would basically explain the issue that they're facing while programming to this duck. Then it actually helps you solve the problem because you're, you're thinking about it like as you're explaining it out loud. So it sounds like you're almost doing that like with yourself , you're like debating with your own voice and then it gives you clarity.

Kazzy (00:06:48) - That's so funny. You wanna know something crazy. Actually, I'll answer you Xin, but I'm gonna answer that. What I used to do in university, I would get so bored doing exams, like three hour exams. So in University Carleton, I studied marketing and uh, um, when I would do these exams, it'd be three hours long and I'd be so bored that I would have to literally, bro, it's gonna sound crazy, but have you guys seen the movie Inside Out by Pixar?

Bobsin (00:07:15) - Mm-hmm. with the feelings and they're all personified.

Kazzy (00:07:17) - Yes, exactly. Literally that Bro

Kazzy (00:07:22) - Just, so I just sit down and do the damn work. I would gamify my experience writing. I would have like five or four characters in my head and I'd be like communicating with them to do the exam and finish it. Otherwise I'd be so bored that I would be communicating with my, like my math part. And then there was another part, it was like artistic part and it was, I can't remember the other ones. And like, that's what I would do. I'd be okay guys. Is it this one? Like, because otherwise I would, I couldn't sit for three hours writing, bro. I can't, I think you now know I have d adhd after I told you both these things.

Bobsin (00:07:57) - Where's the math Kazzy? When do I get to meet him?

Bobsin (00:08:06) - Interesting..

Kazzy (00:08:08) - But yeah, Xin to answer your thing, like, I think I do my best work when I don't hear my voice when I'm not constantly overanalyzing. So when I'm, because when I go back to my voice, I start thinking as cay those same loops thoughts. Like, oh wait, is this right? Is this, I'm like, fuck. Like I don't, I don't want, I don't want, I don't want that to be there anymore. I wanna go back to like feeling like I don't even, uh, it's not even me, it's just, you're in a zone, you know? And I feel like I was listening to a meditation, like a therapist on a video like six months ago talking about when you're most happy it's when you're not thinking of thinking, like when you're playing sports, you're not thinking about stuff. You're so present that you're not, you, you're experiencing like, you don't feel like you're experiencing, you just are experiencing. You don't feel like you are. And so when she said that, I'm like, oh yeah, when you're not thinking, when you're at a concert, you're not thinking you're just present. So when you're, when I'm zoned in and I get to that I this flow, I guess it's called flow, that's my favorite. So it's funny cuz I'm the opposite of that. Like, you're hearing your voices, whereas Yeah, I'm trying to suppress them. Mm-hmm. or it suppresses itself when I'm most focused. If that makes sense. Mm-hmm.

Bobsin (00:09:20) - I think you're more on that side too. Yeah. That's why I need so much music and like, I almost need like, lyrical music, so it's like distracting me in a way so that I can focus on what I'm doing. I don't know. It's seems

Kazzy (00:09:32) - That's wild. But yeah, I can't even have music. You have lyrics?

Bobsin (00:09:36) - I couldn't. Yeah, I do all like lyrical music.

Kazzy (00:09:38) - That's crazy. Wow. Interesting though. Interesting dynamics. But since we're talking about you Bobsin, nice little, uh, segway. So guys, look, any, everybody listening to this podcast, guys and gals? Can't say guys anymore in 2023 without getting canceled, but, um, this is the, yeah, in a previous previous podcast we spoke to Xin. So Xin is the CTO Bobsin's the CPO, like the engineers. And I'm the business side, CEO of our company. Just for context, in case this is the first time you're listening to this, and we spoke to Xin previously about his startup, his journey to getting here. If you wanna listen to that podcast, you could probably see it probably episode five or four or something like that. Go back and listen to that one. But today we wanna talk about Bobsin's origin story, how he became the Dark Knight himself. So how like, yeah, let's just like, just man, you know, you know, it's just casual. So how did you get into it? Like, how'd you first get into coding? Like when was your first line of code written and why did you do it?

Bobsin (00:10:44) - Yeah. Um, I actually got into it I guess somewhat late, especially compared to like people nowadays. But my first line of code was in like grade 13 because I took a victory lap after high school because, I didn't really know what I wanted to get into. And then during that time I was like talking to my dad a lot. Um, and he had like, he still has a lengthy career in software for like, the past couple of decades. So that's what sort of inspired me to look into it. And then, um, I took a couple extra courses, got into like a computer science program and then sort of went from there.

Kazzy (00:11:18) - Not grade 13 a, what do they call those again?

Bobsin (00:11:24) - Victory Lap.

Kazzy (00:11:25) - That they call it Victory Lap. Interesting. Yeah. So, okay. Um, so -

Bobsin (00:11:31) - Just to go back into that actually, so like, that's like the first line of code when I started like actually getting into programming. But prior to that I've always been like really into video games and I would like get really deep into like hacking some video games and stuff too, or like running bots and stuff. Um, because I've been on a PC since I was very young because my dad was so close to the computers. So I feel like that definitely like planted the seeds, but I didn't really start programming until after high school.

Xin (00:12:01) - Definitely. But yeah, I feel like you, um, the seeds that have been planted like this helps a lot.

Bobsin (00:12:10) - Yeah, I've reflected a lot on that because I like distinctly remember, um, a lot of conversations with my dad where he would just be like trying to work on a side project that he's passionate about and he would just like brain dump to me, like all what he is trying to like code basically. And I was, I didn't, wasn't a programmer at the time, so I had no idea what he was talking about, but I was being a good son, just sitting there and listening. Um, and I feel like all those random conversations where I didn't really know what was happening, were all like slowly providing some foundation for me to, to then use later on.

Xin (00:12:45) - For sure. Nice. The subconscious is very powerful.

Bobsin (00:12:47) - Mm-hmm.

Kazzy (00:12:49) - Whoops. Something I just raised my hand. Kaz has something to say. Yes, I do have something to say.

Bobsin (00:12:55) - Kazzy always has something to say.

Kazzy (00:12:57) - Yeah, let's put that down. So, okay. How do you get like your first job as a, in programming coding because like, you can't learn on the job yet, right? You're not gonna learn to code on the job, whereas you could learn all the many, many other jobs you could learn on the job. This one you can't. So like, did you go to school for it? Did you have to teach yourself on YouTube? Did you build stuff mm-hmm. , like how did you do that?

Bobsin (00:13:22) - Yeah, so I went to Ryerson now called Toronto Metropolitan University for computer science. So they like some introductory programming courses there. Like, that's where I started learning some languages. But, um, especially in like first year and, and second year, it's, it's very like foundational and like you're still taking a lot of mathematics courses, physics courses. So I actually did the bulk of my learning on my own. Um, and I was in the co-op program. So my first role was actually like after my first year, um, or after my second year in like that summer. But prior to that, because I was just so interested in, in learning programming, I actually had a couple of side projects under my belt and then one that is not live anymore, but I built this website called techether. That was, uh, it was like an online programming resource tool where people could go in and post, um, like a guide that they used to learn the Java programming language for example. And then you can do like rating systems and stuff. And it was basically just like these Airtable lists, but like almost a decade ago. Um, so that, like, I actually built that and then the, the first person that hired me was like super impressed with that experience cuz I ended up posting that on Reddit and like, it got actually got pretty popular and like this /learnprogramming Reddit. So I think that whole project was, was super important in like me learning about programming.

Kazzy (00:14:54) - Cool. Very cool. And, um, when you were TMU, Toronto, metropolitan University till now, are those, uh, is, has the code or all of that world changed a lot? Or like, is a lot of that applicable today? Like is the logic of coding still what you learned back then? Or is it so different the languages that it's obsolete or it's just useless?

Bobsin (00:15:17) - It's really different. I would say, um, the most useful part is learning the foundations of algorithms. Uh, some database stuff, like some basics of system design. You get that, that from university, but mm-hmm. and, and I guess like the labs where you're actually like building stuff, that's where you get most of the actual real world experience in Univer University. But around that there, there's so much like stuff that is, is really like outdated and as a like computer science student today, you almost are, at least it depends on the degrees. Like I haven't really looked at the most recent curriculums and depends on the school, but I think you're almost like forced to go learn on your own so that you have enough experience to, to really like hit the workplace running when you get your first job.

Xin (00:16:03) - I mean, just -

Bobsin (00:16:03) - Because it's evolving so fast.

Xin (00:16:07) - Yeah. I remember you were, um, last time we were talking, you were saying that, um, you were juggling between, like when you were working, you were already working full-time when you were, uh, studying. So we were kind of like juggling between full-time work and full-time studies to how you manage

Bobsin (00:16:22) - Yeah. So, so, so provide more context there. That first co-op term, um, that was actually at AdParlor and then they liked me so much that they offered me a full-time role. And then I, I basically worked out a deal where I could work full-time at AdParlor while still continuing my full-time, uh, computer science degree. So I was like biking back and forth between work and school and I would really only go for like, mandatory labs and exams and everything else, I would just study on my own time. So I definitely had a non-traditional university experience because of that. But juggling that was definitely like really hard. There were, there were a bunch of times where, um, I had like three years of experience at AdParlor and I was already making like really decent money. Um, and then I would have to like leave my really important job to go do some like elective exam or something.

Bobsin (00:17:19) - It was like a French exam, not like French is bad, but like, like a history exam or some, or something like that, that I didn't really care about. and I got really close to just dropping it. But, um, I knew like as a Canadian, having the bachelor's degree is really important to getting a visa to work in the US so that was actually like one of my main motivators to like continue with my degree, but otherwise I probably just would've dropped out and just like relied on my work experience.

Kazzy (00:17:45) - That's funny. So you'd have to talk to your characters inside out to get the job to get these exams done too, eh?

Bobsin (00:17:53) - Yeah, I remember, uh, I'd go to a lab and then the lab was like build some like basic web app or something, and like at that time, this is what I was doing full, like I was a like a senior software architect at my company. Like I'm leading software teams, like doing this stuff. So then I literally just like built the whole thing in like a couple hours and then I was like doubling it to my, uh, peers in like this lab group. And they were all just like, what do you mean? Like, what is this? And just like the expression from not only the peers, but like the, the TA as well where they're just like, what the hell? Like, it's like a full blown application. It's like a 101 course.

Xin (00:18:31) - What was the app?

Bobsin (00:18:32) - I think it was like, it had to do with bug tracking so people could go in and like log a bug ticket and then it automatically gets sent to the database and then you could like analyze all the bug tickets that are tracked, something like that. But

Kazzy (00:18:46) - It was more advanced than what everybody else is doing.

Bobsin (00:18:49) - Yeah. I just, I, I took it as my own little fun learning project and then I just blew it out of the water

Kazzy (00:18:55) - . That's funny, man. So you aced that class, I'm gonna assume?

Bobsin (00:18:59) - Yep.

Kazzy (00:19:00) - Nice. Okay. AdParlor, like how did you start there? You just applied to it and you just went or you got referred?

Bobsin (00:19:06) - Yeah, so, so that was that co-op position that, and then that hired me full-time. So the way I got that co-op position was like through, uh, Ryerson because I was a part of the co-op program and just applied through there and, and got the, got the role.

Kazzy (00:19:20) - And this is when social media was popping or like just starting?

Bobsin (00:19:24) - Yeah, this was, uh, I guess, no, it was, it was pretty late. It was like 2014, I guess. 2013.

Kazzy (00:19:32) - Not too late. That's not that late.

Bobsin (00:19:35) - 2014. Yeah, I mean -

Kazzy (00:19:36) - Like, that's what -

Bobsin (00:19:37) - So I can share some perspective, like some of the projects that we were doing were like integrating with Pinterest and Snapshot when they were like just rolling out their advertising APIs. So it was, it was, it was late for I guess like Facebook, but it was definitely new in terms of like all the new other social media platforms coming up and just the maturity of the, the space.

Xin (00:20:01) - I remember 2014, that's when Instagram was starting to pop off.

Bobsin (00:20:06) - Yeah, like later on. I, I remember working on a project that we integrated with Facebook at the time and then Facebook acquired Instacart, or sorry, Instagram, um, quickly, uh, like incorporated them into their ads APIs. And then within like a very short timeframe, we were just passing in a new parameter to the Facebook API to then advertise on Instagram. And I was just so impressed by like how fast they, they merged those products. But yeah, you

Kazzy (00:20:37) - Do that often, eh? mix up the words Instagram and Instacart.

Bobsin (00:20:40) - Oh, all the time. Esp there's so many people from Facebook that went to Instacart that it was so funny seeing them do it too.

Kazzy (00:20:46) - That's funny. That's funny. So speaking of Instacart mm-hmm. , how did, how did that come to be?

Bobsin (00:20:54) - Yeah, so I, I left, I basically stayed with AdParlor for almost four years because I had this, this agreement where I was also work, like going to school at the same time. And it was, it was too weird to like get another employer to agree on that. So once I graduated, then I started looking for a new role and one of my previous colleagues at, at AdParlor started working at Instacart and he referred me, and this was when Instacart was just starting their, um, engineering office here in Toronto, which is like, they're, they're second engineering HQ now. Um, so yeah, I got referred from a friend and then, uh, I was there like right when they were starting the Toronto Engineering office.

Kazzy (00:21:36) - Do they still have that office?

Bobsin (00:21:38) - Yeah, well they moved to a bigger one, but, and then now they bought another floor on top of that one. But yeah, I mean they still have the big office here in Toronto.

Kazzy (00:21:48) - How was the difference between going, like, so you were there when they were just starting to blow up?

Bobsin (00:21:53) - So I joined Instacart in 2019, so they were already worth like 7 billion. Um, and they were just going into Canada, so at that point they were still only us. Um, and then around that time they were just going into Canada. They acquired a company called Unata, which was a Toronto based company. Basically acquihired that whole engineering team and then that sort of helped spawn this new engineering hub in Toronto.

Kazzy (00:22:21) - Nice. And, um, YC. So then when did, when did they do YC again? 2011. 2010.

Bobsin (00:22:28) - 2012.

Kazzy (00:22:29) - 2012. Okay. So Instacart, the difference between being an Instacart and AdParlor, what, what was the one main difference?

Bobsin (00:22:38) - Yeah, it was so different. So AdParlor for reference at the largest size of our engineering organization, like engineering and product, it was probably like 25 people or maybe like 30. Um, when I joined Instacart there was probably like already like around 200 engineers. Um, so that was the first, so I only really worked at, AdParlor and Instacart. So that was the first experience of an actual like large tech startup, especially because they're coming from like San Francisco, they have that whole culture that's like ingrained in these valley companies. Um, so it was so cool. Like I still remember just going into work every day just so happy because, um, I always looked up to YC and like the whole ecosystem over there and this was my first chance to actually like, make an impact at one of those companies. Um, so there, there was definitely a bit of adjustment, but the way these YC companies are like, especially at that stage was like, they're still very much startup mentality of shipping fast and like iterating quickly. So at least like that transition wasn't as drastic. Um, it, it's not as like, it's not like Google where you might have to like go to a much slower software release cycle. Instacart was still shipping like really fast at the time, so it was just a matter of working with a lot more people and like so many different software products and understanding how my software product plays inside of like the big picture rather than running like all of the software. Like I did AdParlor.

Xin (00:24:10) - Right. Because not only you witnessed like a jump in team size when you switched company, you also witnessed like the, the, the team at Instacart grow right. Throughout the years.

Bobsin (00:24:21) - Right. Because yeah, like I alluded to earlier, they were just starting that Toronto engineering office. Yeah. So, um, we were quickly hiring across Instacart. Like they would do like a weekly email, like new hires and then every week you'd get a new email of like, at its peak probably like 10 to 20 people that just got hired every week. And like, because we were just ramping up the Toronto office, a lot of them were, were in Toronto. So it was just such an interesting time to be at that company. Um, it's like in terms of like people hiring. Yeah. But then also like, well Covid happened and it was crazy for Instacart because, um, everyone sort of had to rely on grocery delivery because nobody wanted to leave their house. So being at Instacart during that time, I think that was like, it's gonna be a highlight of my career for sure.

Xin (00:25:11) - was it why easy to handle? Oh, sorry, sorry. I would just say why was the, the highlight?

Bobsin (00:25:16) - Yeah. So, um, I don't know. I, I, yes, I think some of these numbers are public, but we essentially like jumped in gross order volume by 5x over a span of like a week and a half, like two weeks. So like first 3x and then like 5x maybe like over like a month. So what that did to all of our databases and our software systems, um, it was pretty like . Some, some databases were essentially like falling over cuz just because like they weren't designed to handle that load. Um, you, you, when you design software, like you have some set of requirements, you ship it for those requirements and then later on, like when the requirements increase, usually you would like prioritize a project to do the necessary things so that your service can handle more traffic. But in this case you didn't have that time.

Bobsin (00:26:04) - Like we only had a couple of weeks to, to really try to scale everything up. So what happened was I essentially acted as like almost like a, a SWAT team where I would like be going on different engineering teams, like wherever people needed help and like scaling their systems. Um, so I got a really cool opportunity to work close with the like core Instacart orders teams, um, and have an impact in like the APIs that are used to create every single Instacart order. Um, and I remember one of the improvements I made during that time was I took off like 13% of all like write volume - which is really expensive - off our main orders database. And it's hard to like explain basically it's like a really good improvement to, to free up some resources for our main orders database. Put in

Kazzy (00:26:54) - Simplest terms, was there, time was, it was crazy, man, you got to experience 5x in a week. Like that must have been so wild. Cause everybody was going crazy that time of the year. People were losing jobs, people were terrified, you guys, it exploded, you know? Yeah. So fascinating. What a time in history. It was a pandemic happening, the world was happening that and you're experiencing what it's like to hyper grow at a YC company. Like it is fascinating as hell.

Bobsin (00:27:22) - Yeah. It was a tough time too though mentally. Like, because yeah, there's also that adjustment of everyone now working remote, right? like you have to deal with that at the same time of trying to just like increase your output of every single engineering team just so you can keep the lights on.

Xin (00:27:38) - Was there a time you wanted to like throw the towel?

Bobsin (00:27:42) - No. Honestly, like, not especially not during that moment. Like, I, I think when, when that was happening you sort of have like the, like holy shit, like , this is actually so cool that you're being a part of it and like you would just, um, absorb all the energy from other teams because, uh, like you would hop on, like I remember jumping on random calls on like Saturdays and on Sundays of like some database data incident. Um, and then it's literally like the CTO and like all the VPs of engineering and like all these super senior staffs, you could just get to hear from them and learn from them. And it was a very, um, like the, the energy was definitely contagious. So that time -

Xin (00:28:26) - And now you are, you're back working with two other person

Bobsin (00:28:29) - Yeah. It's a stark difference, but I love it.

Kazzy (00:28:32) - why, why'd you do that transition? Let's, we don't have to go too deep. I think we talked about it before, but why'd you do that transition from AdParlor startup to a Instacart giant? You grew very like high up there and now back to square one to build, like why?

Bobsin (00:28:47) - Yeah, so that Techether experience in university, um, made me realize I always wanted to like build my own things and I always had side projects that I was working on like throughout my career. So I always had that like urge at the back of my head. I can't really explain why, but I think it has to do with like freedom and like power to control whatever. Like I just wanna like have the, the flexibility to put my energy towards like the things I'm most passionate about. And I think at the latter ends of, of Instacart, um, it was getting to that size where you become so distant from, from like the customer and you just start sort of slowing down. So I wanted to go back to the early days of just shipping super fast, um, putting my energy towards something I have more ownership over and yeah.

Kazzy (00:29:39) - Very cool. Um, I had a question. What was I gonna ask? Uh, it was something about... Xin, jump in if you have anything. I'm trying to remember my question. I was so curious about this one.

Kazzy (00:29:52) - Oh yeah, yeah, yeah, that's right. Okay, so you mentioned, um, you mentioned you went remote and you were in person. So first thing is like, what's the point of like, why wasn't it remote before? If you're just using a computer, isn't it better because you're coding and you don't want people to tap you on the shoulder when you're in, in flow? Like why would you be meeting in person it for this kind of company? Like, I could see marketing or you wanna be collaborative and whatever, but like why wasn't this remote before and what, what are the differences do you think, in that culture?

Bobsin (00:30:27) - Yeah, so I mean, prior to Covid, like our entire company being remote wasn't too common. Uh, we did have cases where like very senior people requested to be remote and it was like a case by case basis, but it just really, really wasn't common at the time. And I, I think the reason behind that is because it's a lot easier to ingrain a consistent culture, um, when you're all like together, especially as you're like growing teams that quick. I think it's so important to be able to like have the people together, um, because they actually get like ingrained in the culture, but also at the same time, like they're more direct in like contributing to it as well. Whereas, um, during Covid at Instacart we were hiring like crazy, um, but we were remote so we had these like really like interesting situations where like, um, people almost, there's like clicks almost , like all the pre remote people and then like all the people that had to onboard like during remote, it was really difficult to to like help bridge that gap and like really bring everyone together because the onboarding experience while you're remote, if it's not handled by the company really well, like it, it's so hard to do effectively, um, because you don't have that like proper buddy system.

Bobsin (00:31:51) - Like you can do remote pair programming, but it's just not as effective as, as being able to like pass by someone on the way to get a coffee. And then you're like, oh, I had this idea and then it spas to this whole like, inspiration or like this whole like new project that should have actually happened, but you don't really have those cycles of like just shooting the shit sometimes. So I can definitely see the benefit to being in person, especially when you're like scaling a team that fast mm-hmm. . Um, but yeah, I I think before covid it, it was like the outliers that would go full remote. So I, I don't, I don't, I I could see why instructor didn't do it before that.

Xin (00:32:27) - So in person is better, obviously better for like the company culture and it's better for like onboarding experience. Would you think, would you say that um, there's a difference between remote and in person for like the growth, uh, environment? Um, is it proper for like junior developers mm-hmm. or people with less experience in general?

Bobsin (00:32:51) - Yeah, that's a great point. I think it's, it's definitely different. Um, I think senior software engineers and above like have the benefit of already knowing how engineering organizations sort of piece together and they know like where they should put their energy, like the right people to talk to, um, just like how things sort of work so they can jump in a new company and, and sort of pattern match and, and get a good sense of like how to be productive in this new environment. Whereas the junior engineers, they hop in and they might not have had any, any work experience before. Mm-hmm. So all they really have is like university experience, which doesn't translate well to, to the job. Um, so they might get assigned like a program, like a onboarding buddy, but like I said, like the, it's just not the same as as being there with somebody and like taking them out for lunch because they're a new hire and everybody gets to meet them then, you know, like, uh, you miss out on all those, those little things. Especially as like a junior engineer where like you're like terrified. Like I remember being so terrified coming into the, like the software industry because it's super intimidating. I mean, um, yeah, like you, you've, you're just coming outta university, you know, you might, it's hard to have that confidence going into a company to to really be able to speak up to, to even know what to ask for. Um, so unless the company is doing a great job of, of like giving you those, uh, boundaries or guidelines and it's really tough for junior engineers now.

Kazzy (00:34:18) - And what would you, uh, and okay, kind of to wrap up in a bit, but like, if we, like what kind of company do you, um, do, would you like for us to build in terms of like the early years? Like we just started working together like two months ago, two and a half, not even. So it's very, very, very early because it built so much and we built so much like good, pretty good traction in that span. So I only can see what would happen in the next six months to a year or two when we grow, especially the engineering team. And because we're in the creative, like in the creative space, like we can also, it's not just engineers, it's gonna be a lot of creative people, people making content, people creating campaigns for companies, all that stuff, right? So it'd be a very fun dynamic environment. Mm-hmm. , do you think you'd like to stay remote or do you think it would be cool to have like that really cool looking office from those movies and shows where you see it, you're like, oh, that, that's dope. There's like an area to sit here with green grass and whatever. Like what do you see more?

Bobsin (00:35:24) - Yeah, well I think there's a hybrid and the benefit that we have now, there's so many, uh, examples that we can copy now because like companies like, yeah. So just to name companies I think is a good, like, framing. It'd be like Linear, like HubSpot, PostHog and Arc (The Browser Company), which I mean, I, I preach all of those examples to you guys, but for listeners you may wanna check it out. So all of these are like heavily product focused, focused companies ship super fast. And some of them are actually like remote by default, but like PostHog for example, I believe they still have two offices or maybe one, um, like basically at their HQ still. Um, so I, I think we could have like an HQ because I think the co-locating is, is super important. Like especially in the early days as we start like hiring others too. Um, but I do love the idea of like globally remote for people that want that option. So I think the hybrid and, and we're, we're seeing a lot of companies take that now, like PostHog, um, like Linear I believe as well. So we'll have the benefit of like seeing how it works for them as well.

Kazzy (00:36:33) - I agree with that. Yeah, I think hybrid would be super cool, especially to have like a really cool office at the very beginning and then yeah. People that are in Vancouver, they're like, look, I wanna work here but I can't move away. We're like, what?

Bobsin (00:36:44) - Right.

Kazzy (00:36:46) - Yeah. Cool, man. Xin, any, last questions before we, uh, wrap up and start hitting those PRs?

Xin (00:36:53) - Yeah, don't you guys want like, um, speaking of freedom of work, don't you guys want like to have like, um, sort of a beach house or a house beside the lake where you can just go work? It's like a cottage, a company cottage.

Kazzy (00:37:07) - I was just thinking of that, like when we were talking, I'm like, hey, wait, Kazzy, like what you're really thinking about is that like why don't you mention it? So you just mentioned it, but Yeah, yeah. Like in California, massive crib. Like yeah, I mean, yeah, maybe. Yes. I've thought of it two minutes ago.

Xin (00:37:26) - We just drop by when we want.

Kazzy (00:37:29) - Yeah, that would be cool. Especially like San Francisco, they have these little houses that look like the Facebook one in the early days. Like that's cute. You know, it's cool. But in California, like in Los Angeles though, they have the really cool places, really cool houses that are not those those vibes. I have no problem go. I would love to go to San Francisco if we wanted to, but because of where all the customers are, the brands, the influencers, creators, like most of 'em are in Los Angeles. So like, I think that that's probably naturally where we might go, but seems like that would probably be the most, like the thing that we may do early on anyway. So I don't even know why I didn't even mention that, but yeah, it seems like the most natural shin's laughing everyone, you guys can't see this, but shit is dying of laughter and muted himself to avoid the, the agony of truth. Right. Shin

Kazzy (00:38:26) - Anyway, so cool guys. What is it? 38 minutes. Amazing. Really, really good. Um, so I hope everybody listening, got to learn a little bit more about, uh, about Anthony, about SSON and about how we, uh, just about everything we did and talked about and how we'd love to grow and what it's like to be at a company that's exploding. Uh, like Bobsin's time at Instacart. And if you're trying to learn how to product, build product or code or anything, build your own stuff. It's that simple. Yeah. And there's a new school by the way, for coding. It's called ChatGPT University. It's gonna make things a lot easier, so, uh, . Yeah, I don't think it's, I don't think it's um, far that everybody can start coding anyway. Yep. Hope you guys enjoyed the, the, the not the video podcast. Make sure to make sure to, uh,

Kazzy (00:39:17) - I didn't say I didn't say it. Make sure fuck. Sorry, I meant make sure to la Yo, I can't man. I'm so used to years of making YouTube videos. Make sure to co wow. I can't even say it. Comments. You can say comments, follow and give us follow. You guys do it. Somebody do it. Man, I can't do this right now. I can't do it, man. Make sure thank us to give us a like, subscribe and follow us for more episodes. You can't do that though. We're not a YouTube video. You see what I mean? Like, you have to like rate us five stars and whatever. Anyway, you know what to do. If you're on a podcast, save us tune in next week and, uh, we'll have, we'll have a cool guest next week for you guys. So make sure to, um, follow the podcast on Spotify, Apple Podcasts and, uh, yeah, time to hit those personal records. See you next week. See you. Yep.

‍