Latest / Code Story | Startup Podcast for CTOs, CEOs and Technical Founders / S12 E24 P2: Why Off-the-Shelf APIs Fail Complex Web Workflows and the Blueprint for High-Velocity Product Engineering with Volodymyr & Vitalii Sydorenko, Co-Founders of Gearheart

Transcript

S12 E24 P2: Why Off-the-Shelf APIs Fail Complex Web Workflows and the Blueprint for High-Velocity Product Engineering with Volodymyr & Vitalii Sydorenko, Co-Founders of Gearheart · Code Story | Startup Podcast for CTOs, CEOs and Technical Founders

• 0:04
  [SPEAKER_01]: This is code story.
• 0:06
  [SPEAKER_01]: A podcast bringing you interviews with tech visionaries.
• 0:09
  [SPEAKER_01]: Six months moonlighting goes.
• 0:10
  [SPEAKER_00]: I was lossing on the backhand.
• 0:12
  [SPEAKER_01]: Who share what it takes to change an industry?
• 0:15
  [SPEAKER_00]: I don't exactly know what to do.
• 0:17
  [SPEAKER_01]: It doesn't go as to get right.
• 0:19
  [SPEAKER_03]: Who built the teams that have their back company, is it's that teams help each other achieve this proud of our team.
• 0:26
  [SPEAKER_01]: Keeping scalability top of mind, all that infrastructure was that they didn't even fighting it as we grew.
• 0:31
  [SPEAKER_01]: Total waste of time.
• 0:32
  [SPEAKER_01]: The stories you don't read in the headlines.
• 0:35
  [SPEAKER_01]: It's not an easy thing to achieve, because the self-invested it off and tried it again.
• 0:39
  [SPEAKER_01]: To ride the ups and downs of the start-up line, need to really want it.
• 0:44
  [SPEAKER_01]: Not just about technology.
• 0:46
  [SPEAKER_01]: All this and more, on code the story.
• 0:49
  [SPEAKER_01]: I'm your host Noah Labhart, and today we return with part two of our conversation with Vladimir and Vitali of Gearhart.
• 1:02
  [SPEAKER_01]: I'm curious about how scaleability for the business and for project delivery is being considered in how you set up your heart and how you started in the early days and I'm also interested in if there has been interesting areas where you've had to fight scale as you've grown.
• 1:20
  [SPEAKER_01]: And I'm not necessarily talking about technology scale for gear heart, more of business and team.
• 1:25
  [SPEAKER_03]: I guess I'll tell a little bit about how I resisted scaling under the telemetallic foundation done.
• 1:32
  [SPEAKER_03]: So, as I told, I wanted the liberty to keep team-solve, and that wasn't national.
• 1:38
  [SPEAKER_03]: I did turn down projects because I didn't want to hire more people because I was comfortable in the size of the team, of how we work, and so I did find that.
• 1:49
  [SPEAKER_03]: So when I just started working as developer my second job as a developer, it was a pretty large company and I was hired for a specific project that was about to start.
• 2:01
  [SPEAKER_03]: And I left the company in about a year and the project still didn't start.
• 2:06
  [SPEAKER_03]: And it's not because something was wrong or anything is just the way company operated.
• 2:10
  [SPEAKER_03]: I was doing something in the middle and just I was looking at the speed of how things
• 2:19
  [SPEAKER_03]: I cannot be in this large company that moves so slow.
• 2:23
  [SPEAKER_03]: So I really wanted to have a small team that moves fast and by attracting those people who are against self motivated self organized and really strong is developed as well.
• 2:34
  [SPEAKER_03]: That's how developed, pretty interesting project without blotting this stuff.
• 2:39
  [SPEAKER_03]: When Instagram was acquired, it only had 13 employees, right?
• 2:43
  [SPEAKER_03]: So I always got this gut feeling that you don't really need a large team to make something interesting and disruptive, and I still convinced them that, right?
• 2:53
  [SPEAKER_03]: So there was this part where I resisted growing the team and complicating our workflows and stuff, and
• 3:01
  [SPEAKER_03]: to just for the sake of growth.
• 3:04
  [SPEAKER_03]: So, there was that part.
• 3:06
  [SPEAKER_03]: Still, eventually, as I told, we do need, like at 40 people, you still need to have at least some level of hierarchy, sensations, reporting, things like that, and that's when I also forget all this kelp, so if you need that part out.
• 3:21
  [SPEAKER_02]: I would say that this mindset of keeping the team small was tested pretty hard in 2022 and a little bit later when a crisis hit.
• 3:32
  [SPEAKER_02]: And we had to build proper business development for Australia.
• 3:36
  [SPEAKER_02]: Unconfortable but valuable.
• 3:38
  [SPEAKER_02]: We learned to find clients without a line on peripherals alone.
• 3:43
  [SPEAKER_02]: We just learned most agencies believe in an island dive.
• 3:50
  [SPEAKER_01]: This will be interesting.
• 3:52
  [SPEAKER_01]: Your brothers, you're working together, you're transforming your heart.
• 3:55
  [SPEAKER_01]: As you step out with balcony, look across all that you've built together.
• 3:58
  [SPEAKER_01]: What do you must proud of?
• 4:01
  [SPEAKER_03]: My song is a bit obvious, but like honestly, I'm pretty proud of the team itself.
• 4:06
  [SPEAKER_03]: My shelf here, I have a small Stimpunk style robot figure that my team gave me for both of the day.
• 4:14
  [SPEAKER_03]: I guess it was a while ago, but I think that was the first time I realized I think I built something that's alive, and I have this robot that reminds me of that, I really like the team itself and how we work.
• 4:25
  [SPEAKER_03]: Now, in terms of like projects and technical things that we've done, I'd add smart speed for sure.
• 4:30
  [SPEAKER_03]: That's the most complicated thing that I've built.
• 4:34
  [SPEAKER_03]: And it's quite high load as well.
• 4:36
  [SPEAKER_03]: Capital one bank is using it, and some other interesting things.
• 4:39
  [SPEAKER_03]: You might want to check it out on their landing page.
• 4:43
  [SPEAKER_03]: So, there are just so many features.
• 4:45
  [SPEAKER_03]: And if you take a look at smart speed, if you're using it, there is a formula engine in there, in particular.
• 4:51
  [SPEAKER_03]: It's the next little spreadsheet formula,
• 4:53
  [SPEAKER_03]: complicated in that it's all programming and it's just no that there are many late evenings with coffee when I was just digging for walks and Tying that that thing to work at scale But there's just one part of it.
• 5:05
  [SPEAKER_03]: There's so many interesting bars to smarts with it.
• 5:08
  [SPEAKER_03]: It's the little I'd probably meet but team as well.
• 5:11
  [SPEAKER_03]: So technically that's the thing for sure
• 5:14
  [SPEAKER_02]: For me, actually, I think that it's something that we are still in the middle right now.
• 5:19
  [SPEAKER_02]: There's a few years that we've made over the last few years.
• 5:23
  [SPEAKER_02]: We're thinking of positioning.
• 5:25
  [SPEAKER_02]: Now we'll start the new branch, putting a gear card itself like a product that needs to be designed and sold.
• 5:33
  [SPEAKER_02]: It's harder and more interesting than anything we've done before.
• 5:37
  [SPEAKER_02]: I can't point to a result yet so I know the quality of thinking that's wanting to eat and that matters to me.
• 5:48
  [SPEAKER_01]: Let's flip the script a little bit and this could be individually or together if you have an example, but I'm curious about a mistake you've made and how you and your team responded to it.
• 6:01
  [SPEAKER_03]: We had fair share of situations when something got broken in production and 3M, and we are figuring it out.
• 6:08
  [SPEAKER_03]: So there were plenty of those situations, and it's kind of part of the job yet, way to learn how to make reliable systems.
• 6:15
  [SPEAKER_03]: I guess there was one moment, like there was one mistake that I'm not proud of out.
• 6:20
  [SPEAKER_03]: So we were building one of our first projects.
• 6:22
  [SPEAKER_03]: We were building that was about pretty high volume of data processing, and I made an architecture for that and we developed it.
• 6:30
  [SPEAKER_03]: And the first version that we did, like it didn't really scale, but we run it, and in like in few days, it was obvious that it's not working.
• 6:38
  [SPEAKER_03]: Like as fast as it should be working, I was like, how did that happen?
• 6:42
  [SPEAKER_03]: And so I realized that I made this design essentially based in my assumptions on how things will behave underlords, and things like I read in books, and just assumed how they will work.
• 6:54
  [SPEAKER_03]: and that was a momentarily, I cannot make guests like that anymore this scale.
• 6:59
  [SPEAKER_03]: So we changed it, so it fixed it, so everything was fine eventually.
• 7:03
  [SPEAKER_03]: We had to redo a good chunk of that part of the project.
• 7:07
  [SPEAKER_03]: We had great relationships with the client, so it wasn't like a huge hit for the business or anything.
• 7:13
  [SPEAKER_03]: It was just the part when I realized that I need to change my way of thinking when I'm making this types of decisions.
• 7:19
  [SPEAKER_03]: And so now it's more about liking proof of concept, collecting metrics, and explaining to myself why exactly I'm making this decisions.
• 7:27
  [SPEAKER_03]: And to make sure that I'm not guessing and assuming all things will work.
• 7:33
  [SPEAKER_02]: We're both focused on doing great work for our clients.
• 7:36
  [SPEAKER_02]: Let's be completely neglected buildings at business itself.
• 7:40
  [SPEAKER_02]: No pipeline, no positioning, no structure on growth.
• 7:45
  [SPEAKER_02]: When the market turns in the start of the 22, we felt it immediately because there was nothing underneath us.
• 7:52
  [SPEAKER_02]: Not nothing, of course, nothing in terms of business development.
• 7:57
  [SPEAKER_02]: The response was to treat your card like a product.
• 8:00
  [SPEAKER_02]: You were building it for the first time.
• 8:01
  [SPEAKER_02]: Run discovery, figure out what we were actually for.
• 8:05
  [SPEAKER_02]: Those are positioning, create a real business development process.
• 8:09
  [SPEAKER_02]: Actually, same things we added by them, all our clients, we did it for ourselves after your influence by the market.
• 8:18
  [SPEAKER_02]: And that shift actually changed a lot.
• 8:28
  [SPEAKER_01]: Okay, let's move into the future.
• 8:30
  [SPEAKER_01]: What does the future look like for GearHard?
• 8:33
  [SPEAKER_01]: Obviously, you've shifted to focus to some new areas.
• 8:36
  [SPEAKER_01]: That's exciting.
• 8:37
  [SPEAKER_01]: You're working together.
• 8:38
  [SPEAKER_01]: What does the future look like for the industry for GearHard and for what you're offering?
• 8:44
  [SPEAKER_03]: essentially with introducing your brand and that's partly explains how we see the process, but so we still continue building more complicated projects as GearHeart, and we're also starting to build these somewhat smaller MVPs using AI, and making them a special with AI working these two directions is quite different approaches.
• 9:07
  [SPEAKER_03]: So when working in more complicated projects, we see that at least at the moment the I cannot replace the developers and just be like an accelerator for them.
• 9:17
  [SPEAKER_03]: Working with the more complex projects, we are looking for ways on how we can speed up our work using AI.
• 9:24
  [SPEAKER_03]: But still maintain the same level of quality as we do there, the same level of complexity of the architecture that we have to work with on those projects.
• 9:34
  [SPEAKER_03]: Now to compromise what we're doing there.
• 9:37
  [SPEAKER_03]: And so that's one way of using AI.
• 9:40
  [SPEAKER_03]: and with teleprojects and butprojects, these MV stages, especially, it is interesting that we can now with the tools we've built.
• 9:48
  [SPEAKER_03]: We can have our managers generate quite a lot of but a lot of code and go through some initial stages of the projects really quite far before developers actually join.
• 9:59
  [SPEAKER_03]: So it's interesting to see
• 10:01
  [SPEAKER_03]: how that will happen, right?
• 10:02
  [SPEAKER_03]: So we have these two direction stats on one hand we have beyond the wall, which is this EI native approach, where we start with this mindset of, okay, we are generating all of it, at least most of it, we're just generating and we're building our processes and propositioned around that.
• 10:17
  [SPEAKER_03]: And we see how that is growing, how we can expand type of tasks and projects that we can build this way.
• 10:23
  [SPEAKER_03]: And on the other side, we have the team that's working on more complicated projects.
• 10:27
  [SPEAKER_03]: There's just no way right now projects of that K-leg smarts we can fit into any models context or anything.
• 10:33
  [SPEAKER_03]: So it's a different approach.
• 10:35
  [SPEAKER_03]: So it's interesting for me to see how that will evolve if the gap will still be there.
• 10:41
  [SPEAKER_03]: If this will become smaller and how we can use the iNOR and our work
• 10:47
  [SPEAKER_02]: Yes, so there are like two directions, both driven by AI, but in different ways.
• 10:52
  [SPEAKER_02]: Before, beyond the wall, the brand grasped in and out for early stage founders.
• 10:57
  [SPEAKER_02]: If now we, I mentioned that we build this MVP site sales already in MVP in two weeks, we ask in our staff a question of what we can do to deliver it like in 48 hours or in 24 hours.
• 11:11
  [SPEAKER_02]: And I think it's more than possible next year or so that's where we are going with it and talking about year files and enterprise create platforms.
• 11:23
  [SPEAKER_02]: I think that AI is becoming a layer that every series past product needs to be ready for.
• 11:30
  [SPEAKER_02]: What is interesting, they all should have this clean data architectural field or APIs, permission models that don't break when AI starts acting on behalf of users.
• 11:43
  [SPEAKER_02]: It's a new version of the same part problem we've always solved.
• 11:47
  [SPEAKER_02]: And what we also want to shoot here and go more and more, how can we automate some development even in this huge products to AI and to have the AI and GenOS AI developers connected to our team.
• 12:05
  [SPEAKER_02]: We have far from there now, but everything is moving so fast that I live again.
• 12:15
  [SPEAKER_02]: and to have this next theme with real developers and AI developers.
• 12:23
  [SPEAKER_01]: Let's move to the last question, Velso.
• 12:25
  [SPEAKER_01]: You're getting on playing, and you're sitting next to a young entrepreneur who's built the next big thing.
• 12:30
  [SPEAKER_01]: They're jazzed about it, they can't wait to show it off to the world, and can't wait to show it off to you right there on the plane.
• 12:35
  [SPEAKER_01]: What advice do you give that person having built?
• 12:38
  [SPEAKER_01]: Gearhard, having built startups, having done all the things that collectively you both have done, what advice do you give that person?
• 12:46
  [SPEAKER_02]: I would generally want to see it that energy is so infectious.
• 12:51
  [SPEAKER_02]: I know it's myself when you are like this your idea you just bring energy in every room.
• 12:57
  [SPEAKER_02]: So I would definitely want to just be near this person just to get some of this energy.
• 13:04
  [SPEAKER_02]: But at some point of course I would ask who is the one person who will pay for this on day one.
• 13:10
  [SPEAKER_02]: Not who's a target partage, just name one specific person, because if you can name them, you're still in love with a DM not the business, and that's okay, just know where you are.
• 13:22
  [SPEAKER_03]: I will just assume as a technical person, that I'm sitting at a plane with a technical founder or a technical co-founder of projects, and he's willing or eager to show me like a code and how it did stuff.
• 13:34
  [SPEAKER_03]: If we were to give some advice to young person at the stage, I think it's important to remember that programming and technology it serves the business, not the other way around.
• 13:44
  [SPEAKER_03]: I think many people in me as well, as we younger, we get to get lost in the art of things and how we build things, how beautiful it is internally, and you forgot about that part.
• 13:56
  [SPEAKER_03]: So
• 13:57
  [SPEAKER_03]: There will be many moments where programming as a card will be in conflict with what business needs right now and it just the business should always win.
• 14:09
  [SPEAKER_03]: And it's really hard sometimes for developers to get to terms with that because we want to build stuff that is beautiful internally, but at the end of the day like business always been.
• 14:19
  [SPEAKER_03]: And it's important to have that mindset going to this business.
• 14:24
  [SPEAKER_01]: Vladimir Natali, thank you for being on the show today.
• 14:27
  [SPEAKER_01]: Thank you for telling the creation story of Gearhart.
• 14:30
  [SPEAKER_01]: Thank you, duh.
• 14:30
  [SPEAKER_03]: Thanks for having us.
• 14:35
  [SPEAKER_01]: In this concludes another chapter of Code Story.
• 14:45
  [SPEAKER_01]: code story is hosted and produced by Noah Labhart.
• 14:48
  [SPEAKER_01]: Be sure to subscribe on Apple podcast, Spotify, or the podcasting app at your choice.
• 14:54
  [SPEAKER_01]: And when you get a chance, leave us a review.
• 14:56
  [SPEAKER_01]: Both things help us out tremendously.
• 15:01
  [SPEAKER_01]: And thanks again for listening.