6 ms·

Hey HN, we are Pierre, Jose Carlos and Antoine, co-founders of Palenca (https://www.palenca.com/ https://www.palenca.com/). We make it easy for companies in Lat

Hey HN, we are Pierre, Jose Carlos and Antoine, co-founders of Palenca (https://www.palenca.com/ https://www.palenca.com/). We make it easy for companies in LatAm to verify employment data and identity, enabling them to run background checks and provide financial services. In emerging markets it’s extremely hard to validate income, identity and behavior as most of the worker's data is fragmented and inaccessible. As a consequence, countries like Mexico encounter poor credit penetration (20%) and high levels of fraud (2x the global level). Moreover, 80% of the transactions in LatAm are in cash. As soon as workers are paid, they are withdrawing all the money out of the ATM to spend it in cash. This means the vast majority of financial and employee information can not be accessed through the bank accounts. We were previously building a lending company and had a hard time validating earnings and behavior that we needed to underwrite credit. We tried to do it first with bank accounts, but the information we were getting was either incomplete or inexistent. To solve that we built integrations with other platforms. Later we realized that the infrastructure we built to get access to the data was way more valuable than the lending business we had before. We integrate directly with payroll systems to provide access to any worker account via our API. We already cover 90% of the Gig Economy in LatAm (e.g: Uber, Rappi, Didi). We are currently expanding to other payroll systems (e.g: Starbucks, Walmart). We are providing 4 types of information: Personal Information (e.g. Full Name, Tax ID, Photo), Profile (e.g. Rating, Acceptance Rate, Lifetime Trips), Earnings and Events (e.g: Trips, Delivery) Our product works the same way as open banking. The end-user provides his credentials and his consent. We get his data on his behalf, and we pass it on to the company that is providing the service (e.g: Lending, Insurance, Recruitment). Right now, our main use cases are lending (cars, motorcycle, unsecured loans), direct deposit switch (change the account in which the users receive their money) background checks (validating quickly the data of a Uber Driver for a new marketplace, checking that they're not fraudulent, already experienced) and insurance (pay per km, check if the driver was taking a trip when an accident occurs). You can check out the demo here: https://www.palenca.com/demo https://www.palenca.com/demo

indiantinker 5y ago
Enhorabuena !! Hopefully, this makes this makes gig economy more equitable for the gig workers.
- pierre62360 5y ago
  Thank you !
abraae 5y ago
If I understand this correctly, your clients are people like lenders (e.g someone who lends money to Uber drivers who want to buy a car) who need assurances about the their customer's status, e.g whether they truly were an Uber driver for the last 2 years. So with your system the driver will supply you with their Uber credentials, and you will log into Ubers system on their behalf, extract key information such as the number of rides they have done, and pass that info back to your client. Like open banking as you say. I hope I have that right, I initially thought that an API for payroll meant that you offered a service that paid people and handled taxes in Latin America. (Just my worldview since we are in the HR business). However I now understand that you are in fact an API to extract data from any number of payroll systems. Just thought I would share that with you as I found it initially confusing, but that's probably on me, and your messaging may well be perfectly on target. Good luck going forward and I hope for your success.
- pierre62360 5y ago
  You are perfectly right, your description is perfectly on point, even though our description confused you at first. To be honest, I'm not too much a fan of Payroll API too, but it seems it is the term that is now coined in the US for what we are doing (see that article from a16z: https://a16z.com/2020/10/20/payroll-apis/ https://a16z.com/2020/10/20/payroll-apis/) If you happen to have an another idea about a very quick way to describe what we are doing (less than 10 words), that would be great !
  - abraae 5y ago
    That is an interesting article indeed. I see that after the headline, they immediately revert to talking about "payroll-connected APIs". They use this phrase 3 or 4 times, then they switch back to "payroll APIs" towards the end. All in all I don't know that "payroll-connected API" is any better for you - quite a mouthful and sounds way more techy. Personally I would like "Payroll verification API" but maybe that's just because we're having this discussion and that fancy will disappear like dust in the wind by the end of the day. It might also preclude you from getting into some of the juicier areas that a16z are talking about (i.e. things that are > just verifiying data), but of course you can always change your tagline :) FYI this area is of interest as we briefly explored automating some nasty govt systems using selenium or similar (aka RPA). We would run the robot in a VM that was torn down at the end of every session for security. We presented the UI to the user via guacamole - then once they had entered their credentials, the robot took over. Interesting but not the path we took in the end.

indiantinker 5y ago

Enhorabuena !! Hopefully, this makes this makes gig economy more equitable for the gig workers.

pierre62360 5y ago
Thank you !

abraae 5y ago

If I understand this correctly, your clients are people like lenders (e.g someone who lends money to Uber drivers who want to buy a car) who need assurances about the their customer's status, e.g whether they truly were an Uber driver for the last 2 years. So with your system the driver will supply you with their Uber credentials, and you will log into Ubers system on their behalf, extract key information such as the number of rides they have done, and pass that info back to your client. Like open banking as you say. I hope I have that right, I initially thought that an API for payroll meant that you offered a service that paid people and handled taxes in Latin America. (Just my worldview since we are in the HR business). However I now understand that you are in fact an API to extract data from any number of payroll systems. Just thought I would share that with you as I found it initially confusing, but that's probably on me, and your messaging may well be perfectly on target. Good luck going forward and I hope for your success.

pierre62360 5y ago
You are perfectly right, your description is perfectly on point, even though our description confused you at first. To be honest, I'm not too much a fan of Payroll API too, but it seems it is the term that is now coined in the US for what we are doing (see that article from a16z: https://a16z.com/2020/10/20/payroll-apis/ https://a16z.com/2020/10/20/payroll-apis/) If you happen to have an another idea about a very quick way to describe what we are doing (less than 10 words), that would be great !
- abraae 5y ago
  That is an interesting article indeed. I see that after the headline, they immediately revert to talking about "payroll-connected APIs". They use this phrase 3 or 4 times, then they switch back to "payroll APIs" towards the end. All in all I don't know that "payroll-connected API" is any better for you - quite a mouthful and sounds way more techy. Personally I would like "Payroll verification API" but maybe that's just because we're having this discussion and that fancy will disappear like dust in the wind by the end of the day. It might also preclude you from getting into some of the juicier areas that a16z are talking about (i.e. things that are > just verifiying data), but of course you can always change your tagline :) FYI this area is of interest as we briefly explored automating some nasty govt systems using selenium or similar (aka RPA). We would run the robot in a VM that was torn down at the end of every session for security. We presented the UI to the user via guacamole - then once they had entered their credentials, the robot took over. Interesting but not the path we took in the end.