Artsy Engineering Radio
Artsy Engineering Radio
44: Hackathon #9
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Artsy hosted our 9th annual(ish) hackathon two weeks ago! Sarah, Pavlos, and Matt sit down to discuss how it went, what we learned, and some of the cool projects that people worked on.
Hello, hello, this is an RFC version of the podcast Request for Comments. Usually in this kind of podcasts, we have three guests, and then we talk about something. In this specific episode, we will talk about the hackathon. We recently had the hackathon at Artsy, and we want to talk about it, talk about how it was organised, how it happened, how the results came out, and everything else. So today, I'm the host Pavlos. And I have with me, Matt, and Sarah, say, Hello, everyone.
Sarah Haq:sure. So Hello, I'm Sarah Haq. We have so many Sarahs in the company, I work in the collector experience team. And I guess my primary focus is our home feed and all the recommendation tiles are on the home feed. So my background is machine learning. I'm one of the few machine learning experts at Artsy, I love art. And I love the project that we're working on collector experience. And I'm gonna share also what we did on the hackathon, which is all related to recommendations, Matt go on I know everybody knows you very well. Please reintroduce yourself.
Matt Dole:I'm Matt Dole. I'm an engineer on the Find and Explore product team at Artsy. And we primarily focus on filtering, searching, stuff like that. And I also helped to organise the hackathon. So we'll talk a bit more about what went into that. What went well, what we do differently next time. And with that, I'll kick it back to Pavlos, to start us off.
Pavlos Vinieratos:Yeah. And let me also quickly introduce myself, I guess, I'm part of the mobile platform team, a new team that we'll be focusing on the mobile infrastructure, I guess, and I'm very excited to work on that thing. So let's, let's I guess we can start with the organisation a little bit. So Matt, you organised this one, along with two other people, if I remember correctly, and then the last one, I think it was two years ago, again, you organised how was it? How was the the whole organisation what was needed?
Matt Dole:Yeah, yeah, I've organised I think the last three now or co-organised. The last two with Justin Bennett used to work at Artsy and this one with Nick Spinazze and Jacob Herrington, who both volunteered to to help out with it. And I would say t hat this is by far the best time I've had organising it, that usually, I think I've kind of ended up trying to take on too much where it's like, oh, like, Okay, I've got to do this, I've got to do this, I've got to do this. I don't know, it's fine. I'll handle it, I'll handle it. And this time, I felt like we actually shared the work much better. So it was like, Okay, you're gonna write that email, and I'll write this email. And I'll reach out to those people. And you'll reach out to these people, because, at least for us, a lot of the work of organising is just kind of making sure that we are kind of in alignment with product and engineering leadership expectations, like, hey, is this a good time to have a hackathon? Are we okay, taking, you know, three, four days out of our regular product work to work on other projects? Is there buy in, you know, are people going to be upset about when we chose to do it or things like that, right. So a lot of it is just kind of like getting, you know, sort of stakeholder management, right? Talking to people making sure that they understand what we're proposing when we're proposing we do it, and that they're on board. And yeah, Jacob and Nick did a really wonderful job contributing. And when we talked about it, we all kind of said, like, none of us felt like, like, we all felt like we weren't doing enough, but it turned out really well. So that means we were doing the right amount, you know, that. We had all kind of like, contributed but none of us had sort of, like had to lead everything. So yeah, yeah. So I would say it went really well. In terms of organising and setup, we kept it pretty lightweight. We basically, you know, like I said, touch base with with leadership on the product team. We sent an email to the whole Artsy team, letting them know that Hackathon was happening and putting out a call for ideas. We created a notion board, big Kanban board to hold those ideas and share that out with the team and ask people to share their ideas there. We hosted a kickoff session where we, you know, kind of just like got the ball rolling, told people what to expect. And a couple of just a couple of notes on the hackathon. And then we did a wrap up presentation at the end, where we got to see all the cool stuff that people that worked on. So it was pretty fun. It was pretty lightweight. And yeah, generally really happy with what we came up with.
Pavlos Vinieratos:Yeah, yeah.
Sarah Haq:You also all did a great work in creating so much energy and momentum. And I think I'm not sure if people are mean people or not Artsy are where this was a hackathon across London, Berlin, and various places in the United States. So it was absolute incredible how you manage everything and create so much energy and momentum. It was really really cool to be part of.
Matt Dole:Awesome.
Pavlos Vinieratos:Yeah, definitely a nice opportunity to work with people that you don't usually work. I remember on the previous hackathon, we had a lot of like single user, or single developer teams, you know, or project whereas this one, I feel like, maybe it was one or two that had one person, everything else was like at least three people or something. And that was really nice. I definitely paired up with different people. Yeah, and then the Organising. That was definitely props to you, and they other two guys, really good job. The last meeting, I guess, where all the demos happen. That's, for me, the most exciting part, and also looking at the at the cards for the ideas. So how for the ideas and the cards, how long do you think makes sense to work, or to wait for ideas to gather before people start acting on this, I guess?
Matt Dole:Well, in this case, we we actually meant to send out an email, like before the winter holidays, because we did this hackathon in early January, and we just like spaced it, and didn't send it out for the holidays, and then instead sent it out like right after. So a little over a week before the hackathon. And honestly, that worked out really well, because I think maybe if we'd sent it before the holidays, people would have like, kind of forgot. And we would have like not really, you know, built up that much momentum. So maybe it was better that we sent it after the holidays. And again, like, maybe maybe a week before the actual hackathon started was when we started collecting ideas, we had like 90 ideas, something like that. So, you know, again, if if anyone listening to this is thinking of hosting a hackathon at another company, I think a couple of really important things that that I would, you know, suggest you emulate. One is trying to get ideas from other teams at the company, right? Don't just make it within your engineering or product organisations. Because, it's really amazing to have input from all over the company, there's a lot of you know, whether it's like, workflow issues that a team has, that you might just never know about, but could be fixed in a day or two, or like big scale ideas that you get exposed to that they've had, but have had no one to pitch to. So that's huge. And then the other thing, this is kind of moving in a different direction. But for us, it was really important to make it clear that Hackathon is optional, right? And that no one is expected to work all night. You know, I think there's this perception of hackathons that's like, it is literally a hack-athon, you're supposed to just like, you know, work for 48 consecutive hours, and then like present, and then you know, and it's like, that's not how we want to run it, right? We don't want anyone to have to spend, you know, their their nights, doing something that is at the end of the day part of their job, right? So we tried to really set expectations, like, Hey, if you don't want to participate in this, don't do it. And that's totally fine. And that is that is beyond fine. It's you know, we encourage you to make the right choice for you. And even if you do participate, like, please do so in a healthy way, don't you know, feel like you have to do so much. Don't take on a lot of stress on yourself. If you're someone who enjoys hacking all night, and then you want to like take a day off next week. Great. Do it, sounds perfect. But don't feel like you have to do that.
Pavlos Vinieratos:Yeah, that totally makes sense. And that kind of ties in with with the themes question I have, I guess. And we can both, or all three of us discuss about this. So I have a question about themes. Like what do you both think the theme for hackathon should be? Should there be a theme? Should it be like Matt said, you know, obviously, not just engineering ideas. But should it be from any other parts of the company? Should it be something about the company? Should it be something completely new? Should it be tech debt like, what should it be? Could it be anything? Well, what do you think, Sarah?
Sarah Haq:Yeah, it's interesting. It's an interesting question. And I think what hackathons what's really amazing is it's not one size fits all. It's it's sort of like what I was saying, like you do, you, you do, whatever, whatever suits you. We have people and Pavlos, You were one of the staying up all night hacking, like you got really into it. And that's you. Whereas I was very clear, like, I don't want to code this hackathon. And I got really lucky with the team that I ended up with, because I was actually working with the marketing team really closely. They'd come up with the idea. And I was working with people completely outside the engineering team. And it was it was really fun. Our idea was creating a visual taste profile. So trying to understand your art collecting behaviour. And I really enjoyed it. And I enjoyed the again, I was able to find the team that worked for me find a project that worked for me collaborate the style at work for me the time zones, and I like that. I think if we put too many rules around hackathon, it takes away the fun and I think if some people like to work on their own, let them work on their own. Some people like to work in teams, let them work in teams, some people want to go let them code so I think the fact that Matt Jacob and Nicholas created so much freedom it was Yeah, it really amazing and I think yeah, we don't want to have too many rules because that makes it seem more and more like work, but go on. What do you think?
Pavlos Vinieratos:What did you work on Matt, did you work on something? Yeah. What do you think?
Matt Dole:Yeah. Well, when it comes to sort of themes and requirements for hackathon, yeah, I agree with keeping it pretty open ended. One idea we've tossed around is having, you know, like, historically, we've done like one hackathon a year, we've tossed around the idea of doing maybe two a year or even one a quarter, but having them be a little more focused, so maybe having one that's for these, like, big picture, try something crazy ideas, who knows, if ever makes it into production, and having one that's more like, hey, we have a bunch of teams who like, you know, need these workflow improvements, or, you know, we could make someone's life so much better, we could save them, you know, weeks of time throughout the course of a year by spending, you know, two days of engineering time, right? So those kinds of like, are there opportunities for us to really, you know, kind of help other teams out, but not big innovative idea isn't necessarily just like, you know, optimizations improvements. So, we haven't acted on that yet. We do one of the pieces of feedback that we've heard a lot in the wake of this hackathon is we need to do this more often. So I think we probably will look to do another one in, you know, maybe six months instead of a year. And I don't know if we'll try to, like, target that. Or if we will just let it be a little bit of everything. But yeah, I also, you know, I do like that there's a lot of sort of creativity and freedom in terms of what you want to work on, and what kinds of ideas we get. And there's not a like feeling of, oh, if you're not doing the most, you know, innovative biggest thing, then you're a failure.
Pavlos Vinieratos:Yeah, I would agree with both of you basically, like it should be, I kind of like the idea of having two hackathons per year one, like focused and one, just whatever. Because I totally think that both are very, very necessary. When when we have your regular sprints and sprint work, it usually comes from, you know, what we expect, for our users what we expect for our clients, for our partners, all that stuff. And usually, there's not enough time to spend some engineering time to do something that a different team needs within the company. And that's definitely a lot, like very, very important. But yeah, then I never want us to lose the fact that like, you can just take whatever you want, and just learn about it while making a fun video, or fun demo or whatever. During the hackathon, I felt that this hackathon was definitely more focused on Artsy things, let's say. And I did see a lot of cards that came from kind of tech debt, I guess, but I'm happy to see that not a lot of people work on these things. Because Exactly, they just tech debt. And it's something that we never had time on. But it's not something that people want to spend their time on, because it's fun. So and I think the previous one we had, which was a while ago, it was a little bit more free on the demos, it was less about Artsy and more about just something that ties in with art in general. But I did enjoy this one really much. Yeah, and I. Yeah, I guess the last thing I can add here is, like Sarah said that I did. I was one of the people that kind of worked a bunch of hours, because I really enjoyed it. And that's how I usually do it with this kind of, you know, events, let's say. And I have to say that I did learn a bunch of stuff, I worked with some NFT trials that we were doing. And I really wanted to learn about that. So it was a great opportunity to work in the company, you know, for that. And make some something nice for a demo. I definitely enjoyed that. I learned a bunch of stuff. Yeah, was really good.
Matt Dole:Yeah, I want to hear more about both of your projects. Sarah, will you maybe dive into to the like art taste quiz you guys are working on?
Sarah Haq:Oh, sure. So this is something that Nate came up with. And the idea was just to have some visual cues to understand what users like. So again, this quiz, believe it or not, can solve lots of different problems. But very simply, we talk a lot about recommendations Artsy recommending artwork, recommending artists doing all sorts of recommendations. What we do when we don't have any data on our users, when users aren't engaging with the app, or the website or the emails, and we have no idea what we could recommend them. So something as simple as a quiz that could just show a series of artworks. And then based on that, we can get some really interesting insights about them. If you haven't done the quiz, please use the quiz. I'll post it up, I'll post it our link so that you should all try it out. And it's just a starting point. So even if we can get users to follow artists, a handful of artists that will help increase retention. So sometimes when we're so focused on the bigger picture, we forgot, I forget how these small things can really make a difference. That's why I'm super grateful for my project that I ended up with. And I got so many learnings as well that I can use for these recommendation models that I'm building for the home feed and I really enjoyed working with the team. And I think we we didn't build up the most flashy quiz, but we built up a really, really useful quiz. And we had Lois a designer who came up with amazing demo and prototype was like, wow, this is incredible. So it was just great seeing the team in action and just realising that we have some incredible people. And yeah, it was just a really great reminder. So please do the quiz. I'll send it afterwards. And then see what sort of art artists and artworks up would be. Like I said, right now, it's very simple. We just had some very basic insights. But they're really, really valuable. And I think this is just a starting point. So I am looking forward to getting picked up.
Matt Dole:If I remember correctly, your kind of demo and proof of concept was like Google Forms quiz, right, that it was just sort of like, how can we, you know, in the spirit of hackathon, how can we build something that will start getting us useful information and, and provide this, you know, a foundation, as opposed to it has to be perfect and finished and complete before we can possibly even think about, you know, trying to use it, right?
Sarah Haq:Honestly, wouldn't have been that complicated to just have this quiz in the app, and then try it out like that. But again, I guess most of the time, we were just brainstorming, discussing and really trying to come up with something really useful. Because like I said, we can solve so many different problems with this, or the problem is quite broad. And we're just trying to find the simplest way something useful, something to get everyone talking. And yeah, I really like this intro approach. And yeah, there's a team of, I think, six or seven of us. And that's what we came up with. So yeah, I know, I came up with the Google Form. I couldn't even fill something cooler than a Google form. I'm not gonna lie. But we had a really amazing demo and the app so that that's, I didn't I could hide the Google for.
Pavlos Vinieratos:If it works, it works. It you know, it's a hackathon. So you do it the easiest way you can do it, right? If it's, if it's a Google form, then so be it. If it's like a text file, sure. If it's, you know, a whole application sure, like whatever makes it easier and more effective to show your demo and see if something is useful or not. Yeah, go for it.
Sarah Haq:But I'm super embarrassed how difficult I found the Google forums are hard. That was my biggest takeaway from the hackathon, like Google Forms it really complicated and it might have been easier to get some help from someone to build a website for but I don't know, that's something that yeah, round two of the hackathon, I would do. But it's the biggest, I guess, challenge we also had, we were on different time zones. And so I was in the Berlin timezone. And the most of the team are in New York timezone. So I could only really work a few hours a day. So I think that's, I guess, one of the biggest challenges that we had, but definitely next time, I will try to pick up more. Yeah, I'll pick up more during the day. But forming the team also takes a bit of time, I don't have something that we want to discuss. But the hackathon project starts, you have an idea, you have to form the team who's doing what, how do you want to do direction, we'll take this idea. And that also takes a day to do and I think I would say we focused on that as well. But it worked out well.
Matt Dole:Yeah, doing that across time zones is is hard as well, because you might be you know, even trying to set up a team when you're excited to get going. And no one else is going to be online for six hours. Right. Like, I'd be curious to hear from both of you. You know, I think you both worked on pretty collaborative projects. How did you go about trying to agree on a concept and what you were going to build? And who was going to do what Pavlos, do you maybe want to you know, touch briefly on what your project was? And then kind of how you approach that team building part.
Pavlos Vinieratos:Yeah, for sure. So for for our team, working with NFTs was completely the other way, we decided instead of like, going the simple way, we just do like the smallest thing we can do, but with the actual technologies and everything. So we, what we were trying to do as like a theme was to see if we can have we can have support for NFTs I guess, in general for artsy.net. So that would be that would mean having a gallery add something for sale or an NFT for sale, having a user or client buy something that was listed as as NFT. And then kind of everything that's behind that. So yeah, the theme was, again, kind of cross cross places. So I was the only one in Europe and the rest of the team was in the US in a few places. And I have to say that yeah, it was, like I said, very collaborative, we did pair of bunch, I would say probably like 40% of the time of these three days was on a zoom call, sharing screen and trying stuff out. So I can because I did have this, you know, morning time where I had 3, 4, 5, 6 hours of just me working and what will happen or what ended up happening was I did a bunch of research for a lot of those things. So you know, researching How smart contracts work how we could add, like, you know, how to mint NFTs even like a lot of those things. And then once I had some basic kind of example, or you know, proof of concept, then I would just kind of make this. And then when the rest of the team would, you know, wake up, we would jump on a call, and kinda, I would show them what I had done, and we are on it, to make it something that actually works, or actually, you know, could be in production at Artsy at some point. And then we had a lot of or they had a lot of time afterwards as well, where, when I wasn't there, they could, you know, keep going with the code, or we had a lot of time to just kind of gather information that the engineers in the team would find useful. So we had to ask people in the team, that we're not engineers, let's say questions about how many have we sold already, without official support, how many galleries have done that? How did they do it? You know, which ways how did they pay? Like, how did the transaction happen? Which order like everything, a bunch of questions like this, and what kind of you know, NFTs and art was sold, so we can have a better understanding of like how this would be implemented. If we implement this. That's kind of how our team worked.
Matt Dole:Sounds like you made really good use of the time difference that you were able to kind of queue up work for them to take on, or they could research stuff that could inform your work. Yeah, and it wasn't like, I think that's the ideal for for cross timezone collaboration where everyone is actually benefiting from having some time for, you know, deep solo work, and then also has time to pair together and make collaboration and make decisions that might affect the rest of that shared time.
Sarah Haq:Pavlos, sorry, you're gonna get mad at me for sharing this, but your team you actually formed before the hackathon started. So you will have a little bit of a head start and a different sort of momentum. Whereas the team I enjoyed is the first time I actually was meeting them. And I think that was that was maybe we needed a day to just, come up with that idea, come up with our working styles. And then because the time zone, we had to delay the sorry to share that. But you, you, you were really organised, you had the idea, you came up with the team, you were the pro hackathon team.
Pavlos Vinieratos:I don't remember who added the card on the board that Matt had, but you know, the moment we saw NFT, you know, a couple of people messaged me, and we're like, Oh, how about this like, should we, Like, yeah, so the team kind of form very early. And then we just found other people to give us information. So yeah, definitely, we were very excited about this. All along. I feel like there's been conversation about NFTs in general, for a while. So it feels like one of those things where a bunch of people just, you know, wanted to work on this, do this, you know, find more information about this. But it just never had time. So it cannot, becomes this inbetween thing where it's like, you know, something crazy, but also useful for us, but also, not really tech debt, But something that has been like in the backlog, I guess, for investigation. So yeah, it was a, I was very excited. And I think the whole team was really excited for this.
Matt Dole:It's a really good learning that you're both sharing here that we maybe could do a better job of helping people form teams or making time for forming teams before, quote, unquote, the start of hackathon. So this is something that we should we should take into account when planning the next one is how can we make it sort of take some of that friction away, or make it possible to have a team have some agreement on what's going to be built so that you don't have to spend a day kind of getting familiar with the parameters before you can actually start building because I totally experienced that as well, where it took us some time to discuss and talk with the people who our work would be affecting and you know, get their perspectives. So yeah, it was definitely a process.
Sarah Haq:From my own personal experience with hackathons. I don't think that's necessary. I think that's part of the fun of the hack that I started thinking about critical, I would just more just I was wanted to highlight Pavlos and his obsession with the hackathons. Like he was the biggest cheerleader, he was the true hackathon player. So I actually more wanted to share the different styles. And like course, it happened organically, that the team was really, you know, really ahead, they got really excited on the topic. And I really don't want to get into the controversy of debate that's happening on NFT. So I'm steering away from using that word. But that is the part of the fun of the hackathon. I think that momentum starts, you shoot this imaginary gun, and it's Go, go go. And part of that is forming the team. And I think that's just something I was just sharing that it takes a little bit of time. And maybe that's why I couldn't build something cooler than a Google forum in the end and you know something really last minute. But that is such an important part of the process. And, Matt, please tell us more about your team as well. But I actually think it's nice to have that time and you have the dedicated time to get to know your team come up with an idea and put something together. And that's why hackathons are so impressive that people build this so quickly, and it's just remarkable everybody does.
Matt Dole:No that's that's a good point too. Definitely. I ended up working on some more kind of like workflow ish stuff that the auction team uses a tool that was built in Artsy kind of an administrative tool that is just pretty out of date doesn't have a lot of functionality. And they end up doing a tonne of stuff in this massive Google Sheet that crashes all the time. And they're accidentally overwriting each other's data and filtering each other's views. And, you know, it's it's kind of a mess. So we, essentially, myself, and Daria and Alican, and Barry all kind of contributed to trying to start rebuilding this in a modern framework that the existing app is, I don't know, probably seven ish years old. And it's like Ruby on Rails, which we still use in lots of places. And it's CoffeeScript, which we don't use in lots of places. So we kind of decided, like, what if we could rebuild this using, you know, next.js? And using that as an excuse to kind of learn more about next js? What the process of building and deploying an app? How would we would deal with authentication, and then how we could also incorporate some of this feedback around kind of product improvements from the auctions team that they're telling us? These are the things that our current app doesn't do, which for us would be really difficult to implement in the current app, because the stack is out of date. So can we modernise the stack? Can we add in these cool new things, the auctions team would make use of you know what's going to happen there. So it was a good experience, it was good to just hear more about their workflow, their process, and to do some learning along the way, I think my personal learning would have been like, I got stuck, just kind of try overthinking, like how to do stuff and should have kind of gotten out of my head and talked to my teammates a little more and gotten their help deciding where I fit into our like team and what we were building. So it's a good learning for me just to you know, be open and be upfront when I'm not sure if I'm working on the right thing. And and get help thinking through stuff like that, because it's always good to have, you know, perspectives, especially from people who are maybe more experienced in this product area, right. I don't work on auction stuff. And I was able to talk to Barry who has worked on auction stuff. And he kind of said, this is I think what you should do, and that helped so much. So yeah, it's always take some time to figure that stuff out.
Pavlos Vinieratos:Yeah, totally makes sense.
Sarah Haq:Sorry, if I'm taking this in a different direction, but something that I did hear a lot. And it's really triggers this feeling of an imposter syndrome, which really shouldn't be the case because hackathons are meant to be fun. And it's all about learning. And it's about being outside of your comfort zone a little bit. You organise this, that's I feel really, I feel sad to hear that. But something that I've seen, not in work related hackathons, but in person, hackathons, pre COVID times where mentors, so you'd have mentors that you could contact about various different questions, but they're always sort of technical or domain mentors. Maybe that could also be the different types of mentors that we have, so that we can discuss the sort of challenges that we face when we have these short timeframes. And we have to build something, because we're also, we all think so differently, we also problem so differently. And that's why I feel hackathons are not one size fits all, because we all we all deal with the challenge very differently. And I think we have to embrace that and get how do we communicate so that people people get different support, as opposed to feeling really deflated? And oh, I wasn't good enough. That's, that's a really unfortunate feeling. It's, it's about Yeah, it's about accomplishing anything. I'm what was really amazing. And the hackathon demos, we even had, what was it, Anson that presented his, sorry, he's not gonna like that. I'm sharing this. But he, yeah, he shared that he project that failed. And I thought that was amazing. Because that's also such a valuable part of the hackathon experience, share how you built something that just didn't work. And I really liked that. And we need to embrace failure more, we strive too much for perfection. Screw that. Let's embrace the failure of hackathon projects as well. And the discomfort and the learning in general, that sorry, my very motivational round those feelings.
Pavlos Vinieratos:Correct. I feel the same, like the demo should be the goal or a goal, right? But it doesn't have to be the thing that you deliver. Like, if you learn something along the way, even if you fail, that's like show us that show was what you learned like show us what you had fun with, like it does. It doesn't have to be, you know, maybe it's wrong to say like no one is paying you to do, you know, someone is paying you to do this. But it is like three days. It's just free. Like, learn something, try something completely different or don't. All fine. So yeah, definitely no pressure.
Sarah Haq:I know we didn't have a competitive hackathon with hackathons where there's been prizes. I've seen the people that bullshit the most. They just spend all their time coming up with an incredible pitch. And because one of the hackathons we also tried to build a smart contract. We stayed up all night, and we were so proud of our achievement, and our demo, unfortunately failed. But the team that won was just someone that had a really cool slide saying they're gonna do this. And that how I learned that hackathons are all about the pitch. It's all about so yes, you could spend three days pitching and preparing the pitch. Sorry, I'm taking the hackathon in a whole other direction now.
Matt Dole:Yeah, that's a good point. I also love to Anson's presentation and just, you know, talking candidly about like that it didn't, you know, they didn't go according to plan. And we still learned a lot along the way. And that's going to inform any future efforts on kind of the stuff that he worked on. So, yeah, I'm a huge fan of, of trying and failing, as opposed to not trying or, you know, instead spending all your effort on a good looking presentation or, or something like that.
Pavlos Vinieratos:Yeah, for sure. And maybe with that, we can go to the next and last question that I have about any favourites, favourites in the demos and, you know, things that you worked on, or things that you show other people worked on. And maybe I can start with this. And I really liked the so one thing, I don't remember all of the teammates, but they made something that they called us, Shazam for art with a really liked, so you could, you know, they made a quick change in the app, the Artsy app, and it would open the the camera, and then you would point it to, you know, any artwork. And then based on some stuff they did, I don't know, in the hackathon, basically would give you the exact artwork or a similar one, or a link to the Artsy website. So basically, you would scan something, and then it would open either the exact artwork online or a similar one. And I thought that was really nice. And I have no idea. Like, even the you know, after they explained it, I still, it still confuses me. And I think it's wonderful. So I really liked it. Yeah. How about you, you two? Did you have any favourites?
Sarah Haq:Oh, don't want to pick favourites. I think honestly, it was the most incredible hour and 15 minutes. And I add that 15 minutes because we went over time, it was so amazing. And it was so much energy and all the demos for such unique qualities on every team. And I was really impressed by seeing some of the project by the Data Platform team and their collaboration with data ops and find and explore as well. And it was just great to see that they don't see I haven't seen a lot of visibility from the data platform team before. So it's really great to see some of the work that they're doing. And also something is, again, we don't think about but duplicating our works that exist. Again, not the most glamorous project, but such an important project. And I was really happy to see that. And I also really loved the I think we were joking about the wiggly graphs. I really liked that as well. So yeah, that was just that was I mean, all of them were incredible. But I'll root for the data platform team. Go on Matt, finish us off.
Matt Dole:Yeah, I fully agree with you that there were so many amazing presentations and projects. Yeah, the the wiggly graphs you mentioned was so much fun. That was from Alex, who kind of like helped created some visualisations for Artsy data about like, okay, so which artists have had shows with other artists, and, you know, kind of six degrees of separation, but with artists, all of these different ways of looking at it, which, you know, was totally mind boggling, because we don't get to see that right. We kind of know that that exists that artists have collaborated with different galleries or, you know, galleries have collaborated on shows like the art world is very collaborative in that way. All of these things are related artists, artworks, galleries, shows, fairs, they all kind of correlate or not correlate. That's the wrong word. But they are all related. And we don't always see that. So getting to have that visualisation was so cool. I'm just scrolling through our slides to see what else there was there was work on, like their performance improvements that you know, shave time off for all of our teams, there was work to make feature flagging and split testing easier. There was work to make it easier for teams to generate, you know, two factor authentication backup codes, there was internationalisation work, which is so cool that, you know, for a quote unquote, global company, we are very, very English centric, we don't do a good job of supporting other languages. So seeing work put towards that was really fantastic. Things like you mentioned, D, duplicate D duplicating artists starting work on a like native seller app so that partners can do things from an app on their phone, instead of having to go to a web page, adding two factor authentication to our mobile app, adding gallery badging to help them identify if they're, you know, black owned women owned LGBTQ owned, like, all of these different things that I think really contribute to making Artsy a better platform in every possible way. Whether that is performance, usability, you know, allowing galleries to kind of like, better present themselves and define who they are on Artsy. Like, there's so many cool things that came out of this. And one thing that I'm really kind of proud of this time around is that I think in the past, we've really struggled knowing how to follow up right that sometimes hackathon projects are finished. They are like done, we shipped it, it's in production, but that's rare. Right. Usually they're like partway or we learned some things. But we failed. Or we started and we built a foundation. But what happens next? And this time around, we've really kind of made an effort to work with our product and engineering leadership and talk about okay, so which of these things can we finish? And should we actually get onto our product roadmap? You know, is it going to be a week of work or six months of work? But like, what do we want to invest in it, because we can see that now that we've had this proof of concept that we've had this demonstration, it's actually going to be really, really huge for us. And so I'm really happy that we're having those conversations this time than in the past, sometimes we finished a hackathon. And it's been like, Well, unless you're going to go finish it in your spare time, it's not going to happen. And this time having that experience of like, we're talking about it, we're making plans for how we can actually get these projects finished or be really clear about like, you know, what, we're not gonna invest in this right now. And that's okay. But, you know, at least you know, and it's not this question of who's who's gonna work on this? Anyone? Anyone? So, you know, yeah, I'm so proud of all of the people who who did projects, I'm also really proud of everyone who chose not to and said, you know, what, this is going to add stress, this is gonna take time that I don't feel like I can devote to it. I, you know, I want to really encourage, again, people to, you know, make the right choices whenever a hackathon comes around. Like you don't have to do it. So yeah, couldn't be more proud of everyone who worked on stuff, and also people who defended their own time and needs and made space for themselves. And I'm really happy that we're talking about where we go next.
Pavlos Vinieratos:Yeah, yeah, talking about where we go next. First of all, thanks for giving a great summary of all of the things that people worked on. But yeah, talking after the hackathon, that's really important. I remember in the previous hackathon, that I worked on two things. And yeah, one turned out that, you know, we really wanted and then we decided to make it a ticket, make it official, you know, spend some time to polish it, make it nice, ship it. And then the other one was clear that, you know, people liked it. But it wasn't, like important enough, let's say at that time, to be worked on to be shipped. And still we haven't shipped it and that's fine. So yeah, it was, yeah, as you said, it was nice to learn about it, try it, see what happens, and then discuss about it. I think that totally makes sense. But can
Sarah Haq:I can I just add sorry that the golden always be
Pavlos Vinieratos:And it was really, really great. Cool, then building something that gets shipped. I think that's really amazing. And that creates so much energy, but the goal should also be Hey, did I work with someone that I don't get to have the opportunity to work with otherwise, you know, you get to meet someone new from the company, you get to build different kind of energy morale again, did I try something that I wouldn't do in my day to day job. So I think all these softer skills are also really important. And that's where encouraging hackathon projects in whatever shape like finds this, volunteer somehow. I don't know. I'm kind of different from that. I want to get everyone to do the next hackathon. Everyone has to volunteer. So my role is a bit different. But I think there's so many great learnings and honestly Matt Nicholas Jacob just absolutely incredible work. And you also had a lot of help from Pavlos, he was your cheerleader the whole time as well. So it's a great, great, great work. So excited my first Artsy hackathon, and I just can't wait for the next one. if we don't have any other questions, I guess, between ourselves then yeah, thanks so much, Matt, for organising and being part of the podcast, of course. And thank you, Sarah for being part of the podcast and taking part and enjoying it. Yeah, it's been the three of us, Matt, Sarah and Pavlos. Thanks for listening.
Sarah Haq:Thank you. Thank you both. It's been a pleasure.
Steve Hicks:Thanks for listening. You can follow us on Twitter at Artsy open source. Keep up with our blog@artsy.github.io This episode was mixed and edited by Alex Higgins. And thank you Eve Essex for our theme music you can find her on all major streaming platforms. Until next time, this is Artsy Engineering Radio.