TalentLayer

All right, we're live. GM, GA, GE, everybody. Welcome to what's going to be another great DevNTell. So if you didn't know, DevNTell is a 30-minute window for builders to showcase something they're passionate about or have been working on in Web3. This could be an awesome project they've been working on, demonstrating unit testing best practices, automation goodies, smart contracts, how to structure a project, etc. Basically, if you've got a passion for something, this is your opportunity to share it with the community. And we are ecstatic to have the TalentLayer team here today: Kirsten and Romain. Welcome, gang! Love to have you on here. Welcome to DevNTell.

Yeah, thanks so much, Narb. Hello, everyone! Hello! Perhaps you would like to start with a brief introduction about yourselves, just like a brief background, how you got into Web3, and how you started TalentLayer.

So I'm Romain, I'm the co-founder and technical lead of TalentLayer. I've been a freelancer for 12 years now. I worked for different companies, I hired people, I hired freelancers, I hired people for companies too. And so that's how, during those years, I just liked working in work environments, in the work domain. And so step by step, I worked in Web2 and then just discovered Web3. And when I discovered Web3, I understood that you can actually code also here as a coder. It just blew my mind. I wanted to just learn a lot about that. And step by step, I wanted to find a way to use really the power of the blockchain and have a meaningful project. And that's how I met Kirsten in last July, just before the hackathon at EthCC in Paris. We met together on Discord, chatted about her idea about TalentLayer, and yeah, just matched really well and decided to start coding the first line during the hackathon.

So, of course, you mentioned that the idea for TalentLayer came maybe like a month or so earlier. So my background, I initially was working in politics for a few years because I'm very passionate about just, you know, helping people try to move up in society and, you know, regardless of what your background is, just, you know, being able to create a good life for yourself. And I thought that the way to do that to help people do that was through politics. So I did that for a few years, but I realized that governments are really corrupt and that's hard to do. And the people that are making the real impact are tech entrepreneurs that are building technologies that make it easier for people to build businesses and live their lives and all of that stuff. So I decided to drop out of politics and teach myself to code and become a freelancer. So I was a freelance dev for a while and started using that to bootstrap my own startups. So I had like a number of unsuccessful like marketplace startups, data tech startups. And eventually, I stumbled into a blockchain governance workshop. So I had known about cryptocurrency for a while, but it wasn't really interesting to me until I realized that you could, like Romain said, build software on it. And it was even more interesting when I realized that it was almost a way of humans organizing themselves in a new way and that there were a lot of similar concepts to what I was dealing with in politics and government. So I ended up basically dropping everything I was doing, diving in super deep, asking the person that hosted the lecture if I could help him with anything. That led to a whole whirlwind of me eventually getting into like full-time blockchain governance consulting for enterprises. And I did that for a while. Ended up getting the startup bug again and building out with some friends an NFT marketplace SDK. Grew that for like a year. And at the beginning of last year, I was like, this is such powerful technology, I have to be using it to solve the biggest problems that I've ever experienced in the world. And both in freelancing as well as then building startups and hiring people, the biggest pain points that I've ever had have been always around like finding work and then finding the right people to hire. And I have always done a lot of this on freelance platforms. And freelance platforms are really awesome because they've been able to make it so that literally anyone anywhere can just teach themselves something and then go build up a reputation over time by doing like smaller tasks, then bigger tasks, getting reviews. So it's like this beautiful sort of egalitarian moment for society and work. But these platforms also hold all of your data and your reputation captive. So when they decide to ban a country of millions of people like they did last year, multiple countries of millions of people, it's the biggest work-related humanitarian disaster. So basically, the original vision for TalentLayer is like how can we make it so that people can preserve their work reputation and, you know, still be able to access these amazing networks of humans to find work with. And that's sort of the vision that I brought to EthCC. I was like, okay, yeah, like this is really cool. Like I need to at least figure out how to build an MVP. So Romain and I and another two devs, Lorenzo and Michael, we just sort of kicked off the dev there and then it spiraled into what it is now. And just to talk a little bit about the evolution of TalentLayer, we realized that the reputation element is powerful, but even more powerful than that is like when you have a low-level protocol that's handling like user profile information, transactions, etc., you're actually able to even out liquidity in the hiring market. Because the biggest issue with these hiring platforms, aside from censorship and all that, is you basically have to create accounts on a bunch of them in order to find enough work. Like when I was a freelancer, I was searching on like five platforms at once and even more when I was hiring. And the reason for this is you very, very rarely, even on the biggest platforms, have the right number of workers and hirers in the right place at the right time. But if you start to sort of break down those silos and have the platforms leveraging this open data infrastructure, then you're able to all of a sudden, like when you have a higher supply of one on one platform and a higher supply of the other side on another, it's like they can find each other. So that's sort of the thesis behind TalentLayer and how we came to it.

It's a very interesting background story and just the fact you went from politics to coding. How long did it take for you to do that switch? Just curious.

Oh God, so you're going to laugh at me because people call me insane. So I taught myself PHP dev with YouTube videos for about four weeks. I quit my job before. So I was like, I'm going to quit and then I'm just going to go hard at this. And then on the fourth week, I started applying for gigs. And I mean, realistically, if you teach yourself like basic PHP, HTML, and CSS, like you can start building websites. So I was just building, you know, like a little bakery website and stuff like that, nothing too complicated at that point, and then eventually moved on to more interesting things once I upskilled. But I basically think that learning through work is really, really powerful and freelance platforms let you do that because the barrier to entry is so low. So that's what I did.

Excellent! Love it. Yeah, love these kinds of stories. So I guess do you both have a PowerPoint or something you're going to go through with us today?

Yeah, sharing screen now. Perfect. So I already gave a little bit of an introduction to TalentLayer, but just to make sure that we understand the key concepts, I will jump into the issues. So, you know, I talked a bit about freelance platforms earlier, but it's really beyond that. Like basically every major hiring platform out there is sort of impacted by the same issues of having the right number of hirers and workers in the same place at the same time. And these platforms have done a really great job getting us to a point where now we can hire people that are outside of our city and even outside of our country on the other side of the world. And that's great. It's made the market more efficient than it was. But unfortunately, like I said before, you have these liquidity issues on these platforms. And it's actually the number one reason why hiring platforms die. It's the chicken and egg problem: not having enough workers and hirers to meet each other's needs. So it causes a ton of issues for the people as well, obviously. Like it means that you have to create accounts on a bunch of different platforms, manage those accounts, and for the workers, it's the worst because you're fragmenting your reputation. So, you know, not only are you trusting these platforms to keep your reviews safe and not ban you, not shadowban you, but also like if you have accounts on many platforms, then you basically have very little transparency for people's full reputations, which makes it harder to find work. And basically with all of this, if you think of like the macro hiring market, there's a ton of similarities to financial markets before DeFi. And the reason why protocols like Uniswap are really powerful is because you have like a lower-level protocol that sits behind dozens, hundreds, and can be even thousands of platforms where you're able to pool liquidity, make this tech composable. You can build really interesting, cool stuff. And the idea behind TalentLayer is why can't you do something similar only for hiring markets? So that's like a very big sort of abstract vision. So how do we actually aim to do that? So TalentLayer is basically this low-level protocol that is designed to sit behind hiring platforms. And the way that this pooling of liquidity is possible is you basically have major backend components for hiring platforms living on-chain or on decentralized storage. And Romain will get more into the specifics of that tech in a moment. But basically, when you have these backend components living at a lower level, you can all of a sudden pool liquidity for user profiles, job posts, and if you're on platform B, if you're searching for JavaScript dev roles, then you'd be able to basically query in results from the protocol, which is pulling that in from any number of other platforms. So it is basically designed to increase the velocity of transactions between hirers and workers. And in the end, we believe that platforms will actually be making more money from this because they're not going to have as many jobs sitting unfulfilled, which is the case now. And we are in a very exciting time right now because we actually just launched to mainnet this month. So like Romain said, we met in July of last year and have been building since then. And we also have one marketplace that's live on us. Hello, WorkX team! So WorkX is basically a marketplace that allows you to do like either full-time or part-time hiring. And we also have a bunch of other teams that are in engineering right now. So basically most of these teams are building like freelance platform-style things. The next one to launch is going to be WorkPod over here, which is a smart contract developer marketplace. Specifically, smart contract auditor marketplace. So there's going to be some fun gigs posted to that very soon, so keep an eye out. And really the reason why we're building this is because, like we said, Romain and I, we're freelancers. We've also built marketplace startups. I have so many scars of trying to find the right tech stack to build marketplaces with and making the wrong decisions, having bad technical debt. So it's like we get you, we get you freelancers, we get you marketplace builders. And it's not just us. We actually have this big open-source community. So all of TalentLayer is 100% open-source, literally everything that we ship. And it's all sort of maintained by our team here as well as a number of other open-source contributors. So if you like the vision, please let us know because we're of course always building stuff. So if you want to help out, let me know. And we also have some awesome advisors. So shoutout to Owen, Scott, Federico, and Mark. These guys have been a big help in getting us to where we are today. So we also have gotten like a little bit of investment. So basically, we have been completely bootstrapped until November of last year, and then we got into the Kleros incubator. So Kleros is a decentralized dispute resolution protocol that we use. And then we also were a part of Filecoin and Outlier Ventures' Basecamp accelerator earlier. And yeah, so that's basically the gist there. And yeah, so core principles that we believe: we believe that work should be uncensorable, we believe that reputations should always be self-owned, and we believe that silos should be broken down. So that's a little bit about us. And now we can jump into the tech side.

So the idea now is to show you a bit how this is architected and what are the key components of TalentLayer. So first, TalentLayer is divided in two parts. You have the core, like the kernel of TalentLayer, which is at the end the smart contracts which are all today deployed on Polygon. And you have then what we call TalentLayer Labs. TalentLayer Labs is all the tools that you need to use TalentLayer and really easily come and integrate TalentLayer. So this is at the end the backend, the contract. The contract is based on six key components: the identity, so it's the profile of a user; the marketplace configuration here is the idea with an NFT to have for a marketplace a way to configure their fee, their dispute resolution strategy, and all this kind of stuff. Then we got the service module here. We got all the workflow from the creation of a jobs to the proposal that you can have behind to the end of the jobs. And this is of course really linked to the escrow contract which will handle all the monetary transaction here with possibility of milestone release, reimbursement, and of course dispute resolution in case something goes wrong. So that's a dedicated contract where we provide a dedicated dispute strategy for the marketplace to handle all of that. And at the end of course when everything goes well, both sides have to review themselves. So we got also this part on the review which is linked to the Soulbound NFT of the identity of the user. And so these are the key components. But what is really important to be easily integrated for a marketplace is to give them all these data easily. That's why we developed this second part, which is the indexer layer. Basically, it's a decentralized database that will get all the data from the events coming from the contracts, but not only. As we are storing a lot of off-chain data through IPFS, it's important to have both of them in the same place. So we are doing this work with The Graph. And this allowed the marketplaces on top to just have an endpoint and use graph to query easily all the data that you can imagine: the users, the services, and also some statistics, which is really important. So on this indexer level is that we start to aggregate some data and give also some statistics for marketplaces. Like that they can, for example, develop a matching algorithm based on this data, or you can imagine a lot of different kinds of middleware that you can develop. And so at the end to make it also really easy for these marketplaces to integrate, we created what we call Indieflow. So it's a ready-to-use forkable open-source code that you can just clone and you have the whole workflow of TalentLayer integrated with the indexer handling the pulling of all the data and with the integration with all the smart contracts. Like that, the idea is that if you have a project or an idea, you can bootstrap really quickly in few days and start just testing your product and focus on what matters for you like your business. So this is basically the big picture of how TalentLayer works. Like we see with Kirsten, the idea is to allow interoperable transactions. So all of these key components are interoperable, and both two marketplaces can exchange together: one can have the worker, another can have the jobs, and we give them all the tools and the financial incentive to do so.

Also to facilitate this kind of transaction, we added also external components. One of them, for example, is the messaging. So here is the Web3 messaging using XMTP, which is also really important. Like that, two platforms use the same messaging system based on the wallet of the user, and like that easily users can talk together before having the final jobs. We plan to integrate a lot of this kind of tool just to facilitate the works for the marketplace. And one of them, another for example, is also the notification part just to have some push when an event happened. So this is a bit of details of how it works more on the data structure. We're not going to go into the details here, but just for you to understand, for example, here for a user on-chain, you have only the minimum important information because on-chain is expensive to store data of course. So what we are doing is storing this minimum viable information on-chain and have the CID, what we call the data URI, which is the URI to IPFS information. And on IPFS we store a JSON with all the information that we need. For example for a user: title, about, all his skills, the country he comes from, the picture, the role he wants, and yeah, for example. And so this pattern is repeated with all our different key components: here the service and proposal, the escrow, and the dispute. All of these schemas can be found easily on the documentation, and if anyone has any question, you can also ask us. Maybe you want to jump on this slide? I know you love it. I love this one so much! So this is a little bit of a complicated slide, we know we need to rework it, but it's basically just to say that like, so even though it's easy to think about TalentLayer as like an infrastructure that you can build hiring platforms on top of, like realistically there's a whole massive amount of like other things that are needed in the ecosystem to help like, you know, really user-friendly, like Web2-native style marketplaces to be created. So, you know, like indexing is one thing. We of course have like our indexer which just, you know, pulls all the information from on-chain. But for example, if someone's trying to build a Web3-native Uber-type application on TalentLayer, maybe a Web3 Uber-type application and you know all the similar applications that they want to be interoperable with, they could need something like an indexer that does KYC on drivers because these applications, if they're serving mainstream audiences, need to know that drivers have the proper credentials. And you know, you can think of similar things for all sorts of different verticals of marketplaces. So there's really just a massive amount of stuff that needs to be built, whether it's like, you know, dispute resolution tooling, indexers, even SDKs in different languages, stuff like that. And then of course, all these amazing like actual platforms that can be created way beyond freelance marketplaces. So yeah, that's a little bit about that. You know, another cool thing with TalentLayer is like because you have the liquidity for hirers and workers living at a lower level, like all of a sudden a group that only has one side of the market, like for example a coding bootcamp, they have a lot of devs but they don't have a lot of hirers. Now it's possible for an organization like that to just whip up an interface where their devs can just search for jobs and they don't have to worry about bootstrapping the other side of the market because the other side of the market's just being pulled in from the protocol level from other platforms. So there's really just a whole lot of possibilities here. Like, you know, we of course like we try to think of things, but really it's like it's just up to your imagination, and we're super excited to see what people, you know, build over the coming months.

Just to finish to make you understand what other features we provide to marketplaces. So we talked about push, Web3 messaging. We got what we call delegation and fees covering. I put them in the same bag because the idea behind it is to provide to the end user a better user experience without all the transaction and Metamask, something more similar to Web2. So like that, marketplaces can easily configure that on their admin, and they can choose to cover fees for users or to use a delegation system to do so. Another cool thing that we developed recently thanks to Sismo is what we call zero-knowledge proof, proof of work here, but real work. The idea is that as a developer, for example, you can have several profiles: a public one, but maybe several anonymous profiles. But you still want to leverage this reputation on your public profile. And with this technology, you can prove that you actually like achieved five Solidity jobs that worth this amount of money without revealing all the data and all the details behind. Another cool thing that we got is just called a signature mechanism. Here is a way for marketplaces to add some middleware before posting on-chain and just a secure way. Like that, some of them have some legal eligibility needs, and they can figure out with that kind of mechanism. And just also to give you the key points of the future of TalentLayer and what do we have in mind. A really cool thing that we are starting to work now is all the private profile data permissioning. So what does mean is that for now we are storing data on IPFS, of course. We will continue to do that, but then the user will have the choice to decide this data is public, this one is not. I just want to allow this data if I work with someone. Then when the job is done, the data is now not displayed, not public for this person. And so we are starting to talk with Lit Protocol, really, really cool protocol to do all of this job. and Lit Protocol will also allow us to do a ton more and yeah, this will come later. And yeah, just some stuff that we have also in mind: all the referral system, maybe you want to say a word on this one?

Sure. So, you know, one of the most important aspects of hiring today is, you know, recruiting, referrals, etc. And with blockchain technology, we have an opportunity to make this a lot more like egalitarian where, you know, if I know someone that's a good fit for a job, I should be able to forward that person to an opportunity and also be able to share in the profit of that. So because you have all transactions happening at the protocol level of TalentLayer, we're looking to build systems to allow both platforms as well as like individual hirers and workers and referrers to make configurations on how fees can be split amongst people that are referring people to jobs. So it's still in the planning stages, we're working on the technicalities there, but super excited to see that live. Yeah, definitely. And so the others, for example, content moderation here. We have in mind, for example, Kleros Curate, which are mechanisms for end users or a community of moderators to help the network flagging bad content and stuff like that. Also in also to have in mind the user experience account abstraction of course will be a key feature. So we want as always to make sure that marketplaces can integrate that really easily. So we're gonna do all the research around it, make it open source, and then it's up to them to integrate that. Also on-ramp/off-ramp is of course a subject to make it even more easy for Web2 marketplaces to integrate because of course we start a bit with the Web3 market, but we want to target as soon as possible all the Web2 markets and any other different market. And yeah, the last thing is about reputation and trust and how we can integrate different partners just to have more data on a user on an address to make it that people can trust more easily together and just continue to work from anywhere in the world basically. I think that's it for the technical side. If you have any more questions.

Yeah, that was fantastic, gang. We do have a question from Nate d3v. Also Nate d3v, shoutout to you for hosting a watch party in the Developer DAO server. Good on you, dude. So Nate asks: how do you approach profile verification as a decentralized protocol? Could that be an issue in the future with more users?

Yep. So at the protocol level, we have been very intentional to not do things like KYC users or even do Sybil proofing, or exactly or rather even to make sure that one user identity is tied to one individual. And the reason for that is we philosophically believe that people should have the right to like separate their identity into different facets of themselves. Like if I am working as, you know, I'm a generalist, right? So I freelance as a creative writer, I freelance as a dev and, you know, all this stuff, it isn't necessarily even make sense logically to have in one profile if you're trying to go for these niche gigs. So we want people to be able to do that also for privacy's sake. And because this needs to be the case at the protocol level, it's actually at the platform level, or the middleware level, that you can take steps to do proof of identity, whether that's KYC, Sybil proof stuff. And there's like a whole lot of different ways that you can go about this. Because TalentLayer is composable with like a bunch of other protocols, you can do anything from, you know, if you're on a platform making sure that all users have, you know, a Gitcoin Passport or have a Proof of Humanity ID, things like that. And a lot of platforms will do stuff like this. A lot of platforms also do decide to do things like, you know, minimum viable know-your-customer type stuff, especially if you're in, you know, heavily regulated jurisdictions. So there's really, you know, a lot that you can decide to do. But we made the decision at the protocol level to really prioritize like, you know, the ability for people to really just have the freedom to display themselves to the world how they want to.

Great answer and great question. Thank you, Nate. I guess one more or a couple questions from me before we end off here. Do you foresee yourselves creating your own talent board, so to speak, a UI that uses your protocol? Or do you foresee yourself more so focusing, at least in the immediate term, on continuing to build out at the protocol level?

Everyone asks us this question, like within the past 24 hours we've gotten it a few times. So we believe that a system like this probably will not work if it originates from someone that is a marketplace themselves, right? And the reason for this is this technology needs to be an open, unbiased protocol in order to gain adoption. Like we cannot be overly aligned with any one specific marketplace and we do not want to build any sort of marketplace as well. So really what our job here is to do is to get the core protocol out there, start decentralizing once we have enough platforms in here to really manage the governance, and then just sort of see it evolve. And the way that we do that is through getting as many platform builders involved as possible. And luckily, like I don't know, like we talk to a lot of other infrastructure founders and sometimes it can be really hard to get people to start building with your toolkits. But we actually haven't had that big of an issue with that. Like we have a bunch of teams that are either currently building or have interest in it. So at least as of right now, we don't think we're going to have to even, you know, come to any sort of decision around that, like do we do this or do we do that because the platforms are coming. So it doesn't make any sense for us to even think about doing that.

Cool, cool. Yeah, totally makes sense. And another question I had is do you foresee yourselves enabling people who go through interviews to be compensated for that? So for example, if a company starts to want to pay people for their time when they come interview, do you foresee yourself enabling a feature at the protocol level for them to distribute their native token say or maybe ETH or Bitcoin or something to them for their time?

We haven't really thought about that before. Feature request! There you go. So my initial gut reaction is like that feels almost like something a platform would probably handle as opposed to being at the protocol level.

I think same, yeah.

Makes sense. I mean it's one of the bigger things, at least I always think about during an interview process, especially in tech. Gotta go through like six hours and multiple rounds and sometimes like 95% of the time people don't get the job, right? So what happens to that six hours of mine? But anyways, yeah, unfortunately we are at time. But final, I guess final final question is what's the best way for people to follow up with the progression of TalentLayer and if they have any questions for you, what's the best medium for them to reach out?

Yep. So I would say best way to follow our updates is probably on Twitter. We also have a newsletter that goes out like once a week. So that's like always a summary of stuff. You can sign up for the newsletter on our website if you just scroll to the very bottom. And you can get our Twitter in the description. And then best way to contact us: you can just feel free to email labs@talentlayer.org and Romain and I will respond.

Excellent! There you have it, gang. And for those of you who have made it this far, now is time for you to claim—sorry Kirsten for it covering your face—but now is time for you to claim your special Kudos for being a participant here on DevNTell. So what you want to do is get out your QR code scanner, that's a regular QR code scanner not one from your wallet, scan this code and claim your Kudos as an attendee for DevNTell today. You will have up to the top of the hour to claim and yeah, that is again: you will have to use a regular QR code scanner, scan this and claim your Kudos.

Can we have one?

Oh, I got a special one for you guys. I'll share with you after, yes. And with that, I just want to wish everybody a very happy Friday, happy weekend. If you're at Consensus, enjoy the conference. And we will see you here next week. All right, gang, all the best!

Great! Thank you so much, Narb. Thanks, everyone! Cheers! Cheers!