Why This YC Startup Is Betting That AI Will Get Things Wrong

what happens when two founders who couldn't find the right co founder

that's the story of Viflow

a YC backed startup that's redefining build with AI

instead of generating messy codes that break under production

Viflow is blending low code and no code tools

in order to help non technical founders

today

we dive into how Alessia and Elia went from co founder dating hell

to building one of the most exciting AI death tools out there

hi I'm Alessia Pacanella

hi I'm Elia and we are the co founders of VibeFlow

and this is founder's emotion

before we get started we have a lofty goal this year of hitting 1,000

subscribers in order to help more people build really great companies

so if you enjoy the content

learn something new the best way to support us is by subscribing

okay let's get into it

so before we dive in in very simple words

what does VibeFlow do so VibeFlow is a

an only one platform that allows

anyone to build production ready apps in no code

all the existing solutions are pretty good at the first generation

but they don't tackle the problem of maintaining

and scaling these applications

which is a need that a lot of non technical people have

and what they do right now is that they continue

like they hire a technical team to continue the development

so this is exactly what we are tackling

the problem that we are tackling is allowing them to maintain

and scale these AI generated apps

so go beyond the first generation

yeah which is super important

and make it way more useful for non

technical founder to actually deploy something

rather than just showing a demo or prototype

that doesn't really work in development

so before VibeFlow you were both separately tinkering with ideas

joining residency trying YC co founder matching

what was that period like

it's a big process

I think it's one of the most important process in a startup

and should not be um

like not a take super seriously

I had the opportunity like last year to join EF on entrepreneur

first in Paris

and this is a program like to do co founder matching platform

so you

you try to match with people and build like introduce your identity

and after you have to uh

see if you match or not on the topic on the professional side

on the personal side

and there it's super funny because they celebrate break up

so it means that when you break because it's like you don't lose time

you need to figure this out super fast

and this is

so this is the process that I have been in and I was trying very hard

like I met I think like 60 people in like three days

and at one point I realized that I couldn't find anyone

that was in the same mentality as me

which is like super hardcore mentality

like I want to work seven days of the week

um I don't want to ask someone to work you know

and super ambitious as well

and I know that Alicia was like this

I will tell you the end of the story but I will let her side first

basically

I was working full time for a company and I always enjoyed doing my

my side projects and I tried to

to work with several people

but none of them was willing to go full time with a startup

so it's really difficult to do a startup part time

I don't think it's possible

um so at some point I just like

I just invent

invested in myself and I took a leap of faith and I quit the job

yeah I tried to do the

the YC co founder matching platform

I tried to talk to the

to these people and they were all amazing people

like the co founder matching platform is really

it's like a dating platform

it's like a tinder or Bumble

when you look for a co founder

you really need to have a

not only a personal connection

but also you need to know how the person works

um it's really hard to do a startup with a stranger

and uh

I tried but at the end it felt odd

we kind of went in this process together

I was in Paris she was in Sri Lanka

but we're doing the exact same thing

and we were calling every day as friends uh

like talking about this experience

and we kind of complain about not finding the right match

and being super motivated

but really not finding the person that you want to be with

at one point we just said

but why are we not brainstorming together at least

so at one point

we decided to brainstorm and we directly end up on an idea

it was not by flow Sparta

but uh we saw that the team was working well

so

you're meeting such a high volume of people in a short amount of time

um what are some kind of advice or learnings that

you had from that experience

in terms of founders that are navigating that process

of how they could narrow down

who to work with who to not very quickly

the co founder is the most important thing

especially in early stage startup

like you need to prioritize your team before anything else

because the idea is gonna change anything

everything is gonna change but the only thing that is not gonna change

so I would say that you should not find uh

only the interest like

you should not to go to someone

because you like the idea that they're working on

I think the right way is the personal fits and the way of working

and the values and after the rest will come together

and so you mentioned this a little bit earlier

but once you teamed up

what was the first few ideas that you guys worked on thought of and um

what happened to them

so I have a background in like computer science and AI

but kind of specialized to biotech

and Ali has a background in like computer graphics and um

like computer vision right

and so we didn't had like one interface

except the fact that we were both very technical to start with right

but we had common interest like a mental health nutrition

a health

so we started in the health area and we looked into a very niche

a problem which was like microbiome

deep tech biotech

so we got some initial traction

at one point we came back to what we know best

so AI and data tools this is really where we have the edge

because I have really a stronger edge in like

AI research and her in software engineering

like she has more than four years of software engineering experience

so we came back and we just tried the tools and we look um

online and we saw so many like

it was at the moment where like Lobal exploded and uh

that there was so many complaints

like everyone was super mad on the social

and we trying to understand what was happening

I love that

and I really like the point that you mentioned in terms of like

testing quickly

breaking up with the idea quickly and not getting too hung

up on things and one thing I wanted to go deeper in is

you mentioned when you go into these user interviews

or potential customer interviews

you don't want to lead with bias

how do you practice that in these conversations

it's a super hard process

I did three entrepreneurship program where they teach you that

so I did ewer where this was my first initiation of that

and you read the mom test

I think it's super famous

and after like

the thing is that it's something that you learn by doing and you

you always do it wrong like

you will always learn something that you do wrong

because it's super hard

and the point is like talk to the right people

so you need to do some work before

so you need to um kind of narrow down the problem

and who can be really your customer to be able

to uh

get the data that you need

otherwise it can be noise

you need to be in their in their

in their shoes by understanding how they approach the problem

but what we saw with Alessia also is that we

we actually went in like hospitals

we went inside and I think

this is the best because nobody is gonna tell you what they like

what they say and what they do is completely different

so you need to go on site to understand what is exactly happening

and you will learn way more by seeing than by um listening

yeah I love that point because everyone emphasizes customer interviews

but if you do them incorrectly

it can actually be more detrimental and good

and that kind of experience of living

the life of your user is actually so important and underrated

you mentioned that you built the first viable platform within

a month so what did that very early version look like

so it was by flow without AI

so basically

we allowed people to import their UIs from other tools like Lovelace

for example and then we allowed them to build the backend manually

with this drag and drop interface uh

but there was no AI involved and then it evolved afterwards

why was it important to have the first version be no AI

and have this kind of low code builder

like what were you trying to test with that first initial version

so we saw online that there were a lot of people that were combining

tools like a Lovelace and Itan and supervisor to create

it is um a prototypes

a functional prototypes with a full backends

the the problem with this solution is that

they have 3 different platforms so imagine 3 different subscriptions

um 3 different platforms where the app lives uh

half of the code base because for example

anything doesn't give you the the code behind the app and uh

it was just not it was like a Frankenstein approach

this is how we call it right

and um

what we did was we tackling this need by providing a one solution

all in one solution where people could just build their backend

and then export the code and have their application

and the own and the application

so we we tried to narrow down the first use case was like um

AI agent integration with the database and then obviously you

I added the AI element and then you launch on Product Hunt

where Product Hunt is where makers

launch new tech products to early adopters

you won product of the day and hit 6,000 users in three weeks

which is incredible awesome

how did you actually land those kind of first few

thousand users we kept launching so we as we said

we launched the first prototype on the internal YC platform um

then we launched with YC then we launched with Product Hunt

and then we launched with Hacker News and on Reddit as well

so it's a continuous process and I think this is the trick

like you don't you're not gonna do one huge launch yeah

you're gonna do maybe one huge launch

but in between

you need a lot of feedback and you need a feedback loop where you um

you improve you keep improving the platform

based on the feedback that you receive

and a lot of feedback at the end comes naturally

you don't even have to look for that

people come to you because they resonate with the solution

they resonate with the with the problem

and then they come to you and they ask you specific features

they kept launching on different platforms

I've heard a couple in terms of a

a couple people's different perspective on like

what they think is the most useful thing for them

by being a part of the Y

network

so it seems like there is just the status batch of being a YC company

there is the network and there's the mentorship

if you have to remove two and only keep one

what do you think was the most important during your time at YC

no for me it's definitely the network

we got the chance to meet incredible people

um that are just over the top and they're also good people

I was really impressed by how YC managed to select very good people

um that are also excellent

and what they do so definitely the network and um

everyone wants to help each other

so that's the biggest value

and the network brings you also a lot of visibility

I will agree with Alicia then also the network

I will say that their logo badge YC is crazy

and I don't know if it's gonna change in the future

so that's why I don't stick with this one

because maybe it's super important early stage

but maybe in the future has less important

what I think that the network is gonna grow with you

and it's always gonna be relevant no matter the stage of

because you are surrounded

so first in the batch you met amazing people that are still

in contact and support you for everything right

so if you want to get feedback

have potential customers understanding how they went to

specific challenge

you have all of that to hands and there's a platform where you can uh

reach out to them and you can um

yeah you can definitely always sell to voice is startups

which is a good a big

so I will say the network brings you far

and the badge maybe is at the beginning

but I will see uh

how much uh

it still has so much weight

yeah for sure

so y Combinator for anyone listening

is the accelerator that backs Silicon Valley's most iconic startups

like Airbnb Stripe and Corn Base

and still kind of one of the highest rank residency

startup school program in the world for early stage startups

so this is kind of more of a point of fascination Elia

you've done a few kind of residency

start of school like how do you compare your experiences

the first one was Ewer Ewer was very early stage

so it changed a lot

so I cannot really say about how is the experience today

but there

I really Learned about the basic of startup and customer discovery

like and I think customer discovery is something that you cannot

you don't learn for example

at etech like

you know technical school right

you really need to know that

this is what you have to do to be able to validate the program

and I think this is super valuable

and I will bring those skills every time

on the other side EF is an amazing program

if you don't have a co founder and you're struggling uh

not finding the right match

because they put you in a room with like

70 people that has different experience

so I think it's a really cool experience and they are all full time

which is something you struggle to find right

and I think EF uh

like nailed that because people are full time and of course

only doing that for two months

so for co founder matching

definitely yes and YC

I think YC is like for sure amazing

I would say that the people you meet there are really

be insane because maybe it's the US

but it feels so like that everything is possible

and this is something that you don't really feel that much in Europe

that you always have like

like I don't know in YC I think we became way more ambitious

way more like you don't see any limiting belief anymore

so I think that YC

with the way it is on in San Francisco also help a lot about uh

like just accelerate 10 times the startup

I love that point cause I definitely resonate

I think there's something about the US

especially Silicon Valley

that makes you think nothing is impossible

and you can really dream the world

that kind of mentality um

is something that is not super widespread in the the

the rest of the globe so you get in and within days

you're moving from Zurich and Paris all the way to um SF

so what did that week look like for you

to be honest I think we so we kinda expected it

I know it's weird

we put so much effort into it that we kinda

I like to say we manifested it yeah

so I like

we both had apartments here and we had to leave the apartments

and I didn't even have time to leave the apartment

because I had too many things there

I don't know if I think about it now it looks crazy

but in the process we really were living in the moment

so we were not thinking about what happens next week

what happens in two weeks

I think we were really giving the experience in the moment and we yeah

we moved to SF we

we started living together in um

working 24 7 at a pace that was like crazy um

but it was just exciting I think the

the only emotion that I can think about

when I think about this period of time is

zero fear a lot of excitement and a lot of motivation

and we didn't apply to YC in the previous startup like the microbiome

deep tech all those kind of things like we

when we pivoted and we knew that was the right problem

we wanted to tackle like for the next 10 years and uh

YC was the obvious next steps

and then I'm curious when you applied to YC

did you guys already build the product and launch yet

or was it still like idea prototype stage

it was um it was

idea prototype stage

in the sense that we had put out a playground that we were testing

with a lot of users uh

so in our network for example

and yeah it was a deployed playground

that's it so we

it's nice because we we could put it in the application and the uh

you know the people the

the partners could test it as well

uh but it was still not a food product

it was just a playground was there a moment

kind of during your time in YC

that really changed the way you thought about building

just launch launch something

it's okay if it has a a a bag

it's not a problem like people that really need the product

they're gonna find find a solution with

for that or they're gonna tell you

and they're gonna scream that they need the bag to be fixed

and it's just a way to iterate with the with the product

so it's perfect you just put it out and you you keep uh

fixing bags and adding features

and I think this is what changed in our

um in our way of working and the mentality that we had

because we were really precise

it's exactly the opposite that you need to do in a startup

you need to be fast and and concentrate and prioritize um

on the most important things based on the feedback

because it's always a feedback from the

from the world and not the feedback from you

the thing is that you already have

you know

what you want to achieve and you have a vision about where you want

to be right so it doesn't change really the directions

but it just accelerated it as you say it right

so you need like for example

you said I'm gonna launch in two months

no they make you launch in two in one month

which is good

because you need to know that you're never gonna feel ready anyway

and you need like every time we're trying to plan the future um

maybe it's gonna be businesses that are gonna use it

maybe it's gonna be consumers and our partner was like

you cannot predict anything

you need data from the real world to be able to uh

understand what you need to build

like usually you launch

and after you understand what you should have built

and this is exactly what you need to do

because you cannot wait that you have the perfect product

in your mind you have a prototype of a product that's kind of works

and that has the vision that you want to express to the world and

and after that you go from there

so yes launching a fast and launching often is the most valuable thing

in early

said shut up

and I think you really need to learn to be scrappy and to be okay

if it's not perfect has to be good enough

so everyone's building in the AI def tools

from kind of household names like Lovable N and n

to kind of the newly released Open AI Agent Kit

to many other startups out there

so where do you see kind of VibeFlow fitting in that landscape

longer term as I briefly mentioned before

we really want to place ourselves in the um

in the long term to allow people to go beyond prototyping

and a lot of the tools that you mentioned right now

stop at prototyping um

so whether that's for AI agents or for a web application development

they still stop

about prototyping

and we want to tackle exactly that because there is no really um

any visual or any IDE for back coding right now

um that resonates with everyone

especially when it comes to um

an opaque area of the back end for example

because what people don't know how to do is how to prompt

um the AI or the agent to uh

modify or debug something that they don't see or

I don't understand

because they don't even know maybe what a backend is right

so that's exactly what we want to

where we want to position ourselves

and then you said that for VibeFlow

AI doesn't really touch the raw code

it builds note based flows that generate it

so why does that distinction matter in terms of

reliability

so for example you are trying with AI to build logic of your app

you need to understand what you want to build

before generating the code

and this is what happens on the first side

you iterate with the AI and you have a visual feedback of what

what is building and once you agree with the first

um logic

you can generate the code and the code is generated deterministically

so you don't like

once you agree with the AI of what you want to build

it works and this is so much faster

so much more robust and you don't have to uh

let the AI generate the code

uh not understanding it uh test a bit

it doesn't work you go in a loop like this

for four days you lost time

you lost money and you're super frustrated

so that's a huge difference

and I think that's also what we brought as developers

to non technical people because when you're a developer

you first think a lot before coding

and AI doesn't do that right now

like AI just cut removed everything

like remove the full folder and just redo it again

so it's and you need

like the rules in coding is that you have to be very minimalist

and you want to change only what is necessary

and also by doing that you learn exactly what people need

so you learn how people think

you learn what they what are their needs and what they need to modify

and what are the most important parameters for them and so on

so this is a really good point

yes I

this brought me back to my traumatic days like learning how to code

um so I actually started coding with o camel out of all languages

so super

and like we had to play around with runtime

and like optimizing space brings a lot of trauma back to me

and with Open AI releasing its agent toolkit

and kind of the movement into making these workflows

and then potentially in the future

ability to create applications

do you see them as future competitors

or kind of playing a part of the same ecosystem

I believe that um

is the best validation that you can have for what you're building

because just imagine like open AI is AI by definition by its core

right and the fact that they put out like a no code builder

it just it's a validation that

AI is not there to be able to generate a agent from scratch

it's not fast enough it has a lot of limitation

and so the fact that they put that for developers

not for non technical people

it's really a proof that those tools are really needed more than ever

and people are visual and needs to have a feedback

so for us it's just a validation

and they focus on um

AI agents for open AI and also uh

for developers right now

because you have no AI trying to help you build those agents

and it's pure focus on the agents

but agents will always be non deterministic

so we don't focus only on agents

we focus on back end and we generate deterministic a code

because for example

if you want to have a workflow like connect multiple apps

you can do this in a non deterministic in a deterministic fashion

and you want to have this reliability

like for example every time that the user connect to your platform

you send them an email you want this to always happen

you don't want an AI say OK

shall I send an email or not

so you always need to have a

those automation like Zapier make anything

they will not disappear and that's where we focus more

and also we focus on back end

which is a very hard problem

that a lot of people are not tackling today

and try to replace it with input

AI chance output yeah

super awesome

what's been the most valuable lesson that you've Learned as founders

throughout this process

you always try to achieve perfection on your product on your own

without being connected

to your customers and you think that after it's gonna be easy

but I don't think after it's gonna be easy either

I mean this also goes along with do things that don't scale

like focus on something that is very like niche and and um

and then you can expand afterwards

like solve a problem for one person yeah

one person that comes back every day and loves you

and loves your product and loves your company

and then you can scale after that

but find that one person

I would say that the environment matters a lot as well

I think San Francisco is a big one for that in San Francisco

you never gonna be good enough

which is good

because you always want to push the limits of what you're doing and

and you also like and in Europe

I think you have the

you can be good enough very fast and you don't want that

you always want to like you had goals and they are here

the goals now every time

and I think this is really important to keep pushing

and also iterating fast like be don't do a roadmap for one year

do a roadmap for two weeks and make sure that you achieve the goals

and also be way more ambitious on the goals

yeah I love that my personal motto is if you're dreaming

you might as well dream all the way up to the universe

and then land somewhere in the clouds

why she tells you all the time is that during the batch

at least is like if you don't talk to users or you're not building

what are you doing and it's good

for example you're out in a party and you think about that and like

I know I know I'm not good

I shall go back to building

yeah there

there's so many things that can distract you

and focus is so important because time is finite right

how we like to end this is with a few rapid fires

so would you rather do a demo day redo or product hunt replay

I've done a day review

yeah product hands

ha ha replay

ooh we're seeing some discrepancy

see here haha

harder to find product market fit or the right co founder

right co founder

yeah right

co founder hmm

Zurich or San Francisco

San Francisco

and maybe this one is a hot take

but do you agree building a company in Europe is building in hard mode

for me 100% agree with that

yeah you have to put on your side and you maximize the chance

like the surface of luck I guess

like what do you think are the main reasons

for me the main reasons are 1 network

like in San Francisco

you have to make zero effort to meet anyone you want

like you are you literally meet anyone without any effort

and the barrier to meet them is non existing

while in Europe it's of course it's higher events as well

and then there is the part that

I mentioned before which is the

the the belief that you can do anything so in San Francisco

everyone is like

you can do anything and every every time you talk about something

everyone is hyping you up for example

I don't know like you you raise a 10 million there

it's a it's normal right

you don't think that you achieved anything in Europe

maybe you feel you achieved already

and so you always have the imposter syndrome

maybe like you you need to do more

you need to do way more

you need to do way more because you're not there yet

and I think this is the mentality that you want to have

you don't want to um like be less hungry

you always want to achieve more because you feel that you are

you are a nerd and you being judged and when I arrived at etech

everyone was in the same bubble like they

you being judged if you don't work and I think in SF is the same

if you if you don't work on your startup and they

they're gonna judge you more than here

they might say go take a break go out

it's gonna be fine this kind of things

you always want to be surrounded by people that understand you

and push you to do more

yeah I love that um

if you're not talking to customers or building

what are you doing right

that's a wrap if you like this episode

please hit the subscribe button

it helps us bring on more awesome guests

level up production and drop new series you'll wanna watch

see you in 2 weeks