Deep Dive
Shownotes Transcript
Welcome to another episode of Shuang Cheng Ji. Today, we are honored to have Jian Ran, a friend of mine from Rice University. She graduated in summer 2024 with a major in statistics and has been serving as the CTO of her startup Orbit, where she and her co-founder are creating an embedded financing solution for small and mid-sized businesses.
During today's episode, you'll hear that we touched upon topics such as Jian's gap year before starting college, why she picked B2B FinTech as the vertical to build her startup in, how AI caused the so-called talent collapse, and what is the so-called minimum viable founder. Let's get started. Hey, Jian. Welcome to the podcast. As a kickoff, can you briefly introduce yourself?
Hello, hello everyone. My name is Jian. I'm co-founder and CTO of Orbit. I'm very glad to join the podcast today to share my experience. Let us just begin. So what were you doing before founding Orbit? I started Orbit right after college because I feel like it's a once in a generation opportunity given this huge AI boom. So before starting my startup, I actually tried many different things during college.
So I took a gap year and worked at an education startup company for a year as a product manager, and then tried a bunch of different things like software engineering, both tech companies like VMware and research labs, like the Texas Medical Center.
So it's kind of a mix of product, engineering and research experiences. Got it. And how does that sort of impact you, that gap here? And how does that impact your career? Actually, my gap here is
kind of like an unexpected experience. I didn't plan to take a gap year. I chose to take a gap year because of COVID. So I didn't want to do Zoom University online. Instead, I want to have some actual practical experience. So I did my gap year during COVID and that was a great opportunity for me to actually reflect on who I am as a person and also
understand how the incentives are structured within a society. So how teams operate and how products get built. So just for context for our audience, before you started college and right after you graduated from your high school, right? Yes. So I took a gap year after high school due to COVID.
Okay. Can you talk to us about what did you do at that time? I recall that, like, I read an article up at that time and, like, saying that you moved from a, your hometown to, I think, Shanghai. And then, like, you basically, like, um,
It's pretty entrepreneurial because like you mentioned, you just grabbed your suitcase and then to Shanghai alone and then managed to get through all the hardships, financial difficulties. And then how do you actually start some events and other challenges? Could you briefly walk us through what did you do at the time?
Yeah, for sure. So I would say the gap year is kind of like my first entrepreneurial journey because my parents were not that supportive for me to taking a gap year. They feel like it's like you're delaying your study, you're going to graduate late, things like that. So I actually want to some sort of prove myself the ability to survive in like a big city like Shanghai and actually
I did a lot of creative hacking, not in terms of technical hacking, but the way I managed to bootstrap or leverage resources. For example, I tried to save money on my rent, so I was able to create a co-living space.
And then I subleased this to other international students. So I was able to manage the whole floor, kind of like a co-living space. And then being able to kind of like arbitrage that differences for my own rent.
because I was not get paid for like a high salary working in the ad tech startup. I mean, for a high school graduate, like nobody expect you to like know anything or expect you to like worth a very high salary. So I just like, okay, I got the opportunity and I want to actually learn as much as I can. And I want to survive. And I want to actually, you know,
make friends and then find mentors. I find it very challenging and interesting. So when I come to the United States, I feel like it's the same process, just in a different country.
And it's the same thing for even starting a startup. So, for example, I'm building in B2B fintech space. I don't have the fintech experience, actually, but I was able to find whatever resources I can to actually learn all the process and find whatever resource I need to actually get things done.
So I would say the gap year is a very, very unique experience before I go into college. Great, great. Maybe let's talk about your university experience a little bit. So after you come to Rice, what did you do? I guess like...
maintained your motivation to still do a startup after graduation? Because I imagine that like interning at a startup versus actually starting your own startup, there's still a gap, right? Like keep you, keep your dream of still starting a startup after graduation. Okay. So yeah,
It's kind of interesting. I remember the first, my freshman year at Rice University, I think is the second semester that GPT-3 just got launched.
And I remember... You mean GBT 3.5 or 3? I think 3.0, like the very first GBT that actually sent, not just like GBT 2, that like open source. But GBT 3, like chatting with the chat, it became very clear to me that this is the paradigm shift for like our generation, like compared to previously, like the internet, the mobile, the cloud, like this thing, even though at the time it's not...
working very hallucinates all the capabilities, like even generate like some basic code, I realized that I have to rethink about my education in universities. Because this technology is evolving so fast that after my graduation, a lot of jobs may already be automated. Like even though like, like software engineers, somebody may think they are like, you know, white collar, very high skilled
But right now, all the coding copilots can just generate code in a very fast way. All human beings can write it like that. I actually changed my perspective quite a bit. And also my gap here, I would say it's like my first year in university, but just like social university instead of like a traditional academic institute. That experience gives me the opportunity to talk to many different, like people from many different backgrounds, not just like professionals.
privileged kids who can get into good college, but the people who don't even go to college or people who do more manual labor kind of work. So that makes me think a lot about how I want to create value for society and what do I expect from a higher education. So from that being said,
eventually chose statistics as my major instead of computer science or some other popular major at the time because I wanted to focus more on mathematics. So because at the end of the day, LMs or AI, they are probability at scale. So all the underlying mathematics are actually, I think, the core of how this AI works in the first place.
And also I chose a bunch of different electives like cognitive psychology, politics, religion, electrical engineering, computer science. It's just a bunch of different things that I want to pursue or interested in. So actually a couple of days ago, I looked at the Rice University website and realized they just have a new AI major. I feel like all of my courses
Kind of aligned with the AI major. It's mostly, you know, math, statistics, like computer science, you know, some humanities. That basically summarized the experience. So I was able to leverage AI to accelerate my coursework. So I graduated within three years. So I actually didn't waste as much time as my parents were worrying about.
So now let's move on a bit over the university and go on to your startup. So why did you decide to build a startup right after your graduation? Yeah, so I chose to build up the startup because I think it's a unique window when this AI is so new. All the existing players, they are all trying to figure out how to build for the AI generation.
And for me or for many audiences who are listening to this, we are kind of like, I would describe it as AI native generation. So we are already used to coding or working with AI all the time from day one, instead of trying to learn AI as a skill for many existing players or generations. So I think that is first. And second, choose to start a company right after graduation
like would be considered as quite risky for many people. But for me, I think that is an opportunity, especially given that I would like to pursue my entrepreneurship dream in United States. So as some other international students or founders want to start a company in the US, they have to face some immigration and visa issues.
So instead of trying to get to Google, the fan companies, and then get the H1B lottery, I wanted to go straight to Extraordinary Ability 01 as a founder.
That actually gives me a very intense timeline or eventual goal to strive for. So I know that, okay, I want to become an extraordinary founder. And then here are the criteria I need to meet. And then I motivate myself every day to strive for that criteria. So I think it's either a risky move or like if it's not a risky move, if you actually prepare and make a plan,
want to start a company without no plan and just hope it will work, it wouldn't. So I would definitely say it's not for everyone, but hopefully it will work for my case. I think you've sort of mentioned this point repetitively that, you know, you see when you see ChaiGPT-3, you realize there's a paradigm shift and it provides opportunities for building startups. And so I'm a bit curious, how did you perceive this paradigm shift? Opportunity is it providing for, you know, when you were deciding to build your own startup?
Well, I think I can explain from two levels. The first level is the labor structure change. So before AI, to start a company, if you want to build a basic SaaS company, you probably need a designer, a front-end developer, a back-end developer, at least three people or four people to actually build a functional product.
And you rarely find a person who have all the skill sets that you need. So you need to hire and find those people who can provide the work for you. However, for AI, the differentiation is that it's actually starting to replace labor. So it's able to provide some insights or knowledge or tools for somebody to pick up a new skill extremely quickly.
So for me, I don't need to hire those three people to build the stuff for me. I can just go and learn it. And then there are all kinds of tools for me to help me to design, to build, and do everything I need to launch the product. So from the labor perspective, the good in a product has dramatically reduced because of AI. So that is the first is...
the AI sends a more dynamic product experience. Before, whenever you go to a website or product, specifically software products, not just hardware or other more deep-type kind of product,
But specifically software products, you basically have a fixed set of parameters or pages and then users just click on the buttons and then finish a workflow or something like that. But with AI, it was able to intake more dynamic data. So users' behavior will change dramatically.
So instead of you go search on Google right now, you just ask GPT to provide an answer. The user's behavior changed in a dramatic way that existing companies have to think about how do they provide better services instead of just traditional software interaction experiences. So I think two perspectives, the labor and also the consumer or the user's demand.
So that is the two main reasons that I think this is a great opportunity to found a startup. And just to dig a bit deeper on it, as you've mentioned, sort of now to build a company, you don't need several roles, but the founder can take several roles simultaneously with the help of AI.
What are the minimum capabilities that the founder required in the sense of AI to build a company? So the minimum, let's say the minimum viable founder, put it that way. Yes. So the minimum viable founder, I think is a great concept or like a verb, MVF, which
So actually, I think I would use myself and my co-founder as two examples. So a person definitely cannot do everything. Like that is for sure. We have limited time. But compared to before, when somebody say, okay, you need to be a T-shaped kind of person. You are broad enough and you have a one specific like
like focus. But now we are able to become a pillar, multi-pillar ships kind of talent. So I have some basic, like broad understanding of skill sets and have multiple verticals like software engineering, data science.
However, for me, because I am more of an introvert and I prefer or I enjoy doing product design or engineering more than I enjoy talking to people or do marketing and stuff. So I find the things that I'm good at and the things that I enjoy to be the center
of like the pillar shape I want to focus. So I basically define a minimal viable founder in the engineering space or the product side or a technical founder. I need to at least, you know, do the front-end, back-end data, you know, all these things define what skill sets I need.
Whereas for my co-founder, he's more of a business person. So he is able to do business sales, marketing, design. He's not a great designer, but he's able to produce amazing design just like with AI or help him to do a great pitch with the help of AI strategy. So I think it's very up to the founder to determine what they are good at, what they're enjoying. And then
combine this to basically, I think there's a Japanese word called Ikigai. I think that's the word. It's the center of what you like to do, enjoy to do, and what you're good at, and what the world needs. And then you find that center, then that can be a great leverage point for you to be the minimal viable founder.
Great. Yeah, Rich and dig deeper on the first point. I actually want to dig deeper on the second point you mentioned. Can we hypothesize that as you mentioned, like user experience is rapidly changing because of AI.
that means the existing incumbent that are out there need to adapt really quickly to this change. But because of the nature of big corporation, they are slow to change. And so under this AI age,
be a massive opportunities for startup founders to come in because they can just start building, having in mind of this already shifted user experience. And then you start building specifically to those like already changed user experiences. And because a lot of big companies cannot maybe catch up in enough time, then you can like just take those market shifts. Yes. So
- Well, actually I think this is more of a question about who can serve
the customers better. So for, especially for the B2B case, so currently all the existing products, whatever for sales or like engineering, et cetera, for example, Salesforce is kind of like a one size fit L kind of like solutions because previously you cannot afford to build customized solutions for each unique use case or in specific industries.
But recently, because of AI, there are some very niche AI SaaS company coming out or whatever AI power companies popping up. For example, there's a
a Y Combinator company who is building a SaaS for dental clinics. So it's a very niche industry, but they are able to find the pain points for a dental clinic, like to automate their back office day-to-day work and then build all the products just to tailor to make the user's life easier.
So all the existing companies, they wouldn't be flexible enough to just serve a small niche of dental clinics, but there are more startups who are able to find a specific niche and then just build the vertical stack on top of it. So there's actually a great video on YC talking about the multi-billion or trillion dollar opportunity for vertical B2B SaaS startups.
And not just like B2B, SaaS, but consumer as well. So the business at the end are all, you know, consumers. It's like a bunch of consumers, they choose to start a business. So eventually it's all up to the individuals, you know, choose what they want or what product they want. Also, as more and more people are used to this, you know, chat with AI and get instant feedback instead of click on a bunch of blue links on Google, they will expect the same from customers.
all the products they use. So this paradigm shift is like a trend, it's just a matter of time. Yeah. Maybe we can shift a bit to talk about your current company Orbit.
So maybe give us an overview of what your company is about and what problem are you solving? Okay. Orbit is a fintech startup providing flexible payment and financing solutions for SMBs, so small and medium-sized businesses. So SMBs, for example, restaurants or cleaning service companies or dental clinics, they actually, it's very hard to operate an SMB because they
On a daily basis, you struggle to make ends meet. You want to find a customer, serve the customers, pay your employees, pay all the invoices. So many times there are cash flow gaps in the day-to-day operation of a business. Especially during COVID, a lot of businesses shut down because their cash flow dried up. And those SMBs, they are underserved by the current traditional finance industry.
So the reason is that the SMB is kind of a unique species. So they're not like consumers. The consumers are relatively straightforward. You can just categorize a consumer with a standardized credit score. So your credit score is good, bad, medium. It's relatively straightforward.
Also, it's not like big companies. Big companies, they have pretty standardized or established financial track record. Even though they're a couple months didn't go well, this can still last for quite a bit of some time because of the history and their existing assets. But they wouldn't lend that much, and it's not as straightforward as consumer. It's kind of like an awkward position in the middle.
So currently a lot of SMB loans are served by the community banks in the United States. So they understand the local market as well.
They are able to provide the financing, but the existing providers, your technology and their process is relatively manual and outdated. So SMBs still need more straightforward way to get financing when they need it and then in a very affordable and easy method.
So you're just trying to be like the online version of those community banks across the United States, providing SMBs with easy and quick solutions?
No, not really. Existing financial products, their mindset is very much like, okay, I build a product or I am a bank and you come to us. When you need the finance, you come to us. But Orbit, we are aiming to become an embedded finance kind of fintech companies.
At the end of the day, an SMB wants to have financing not because they just want to finance because they want to buy equipment, or they want to pay their employees or they want to do something like a means to the end. It's not like, okay, I just need financing. What we want is to kind of embed ourselves our financing capability in all kinds of tools those SMBs already using.
Instead of those SMBs trying to, okay, find another tool or go to another person, try to get financing, they can just get it whenever they need it within their platform. So for example, a restaurant owner who are already using a restaurant SaaS product, and they can just like go to that SaaS product, who embed Orbit within it, and they can just get financing right away. So that's why I think the A6&Z general partner, Angela Strange,
mentioned like a couple years ago that every company would become a fintech company so it's not just like okay i would be a define myself as just a fintech company only do financial services but every businesses would touch transactions in some point in their life cycle so for example uber
They have embedded finance to provide instant payouts for driver. So Uber is not a FinTech company per se, but they embed FinTech capabilities in it. From that being said, currently all those embedded finance providers are those big companies like Uber, Stripe, Shopify, Square. They're able to inject some financing capability into their existing products.
However, it's kind of hard for medium-sized SaaS companies or this new wave of AI native SaaS company to embed financing for their customers. So what Orbit wants to do is to become this, the go-to embedded finance provider for anybody who wants to provide financing for SMBs. So that's the overall logic.
So I'm a bit curious, you've mentioned that currently the most of the embedded finance are done by big companies like Uber. So what made you motivated to build a startup in this field where it's dominated by big players? Okay, so I wouldn't say it's dominated by the big players. It's more like a feature that only big companies have. So because currently when you want to...
like create a financing capability within your product, you need to have like specialized engineers know how to build FinTech related features in your company. And for most, most companies is just not realistic to have that, to hire that engineer, build all the stuff and then make sure it works, make sure it compliant. So it creates an opportunity
for us to become this like a widget or a plugin, just become a plug and play kind of like a toolbox for all our partners and
businesses who want to have these features. It's like a modularized Lego block. So you don't need to hire all the engineers to build that in-house. You can just use Orbit. And Orbit already makes sure that we are able to provide businesses with the financing they need, with all the compliance, with all the legal stuff required. So it's very like a Lego piece that you can just put in your existing workflow or your business.
I would imagine the data you've been collecting from your partners, when comparing to traditional, say, fintech data being collected by finance industries, it will be much less structured and much less organized because it was not designed for fintech, this data. Was it an issue for you and how do you tackle it? Well, I think that is actually one of the big advantages we have
because SMB underwriting, as I mentioned before, is very different from a consumer versus like big company. So there is no one size fit all
a credit score for SMB. Even though there are some credit bureaus like Experian try to give a SMB credit score, but it's much better to underwrite them on a cash flow basis. Because the most straightforward way to know if a business is in a financially healthy status is just to look at their bank account, their cash flow every day, right?
So we use Plaid as our integration partner to get the transaction data. It's like the single source of truth for the transaction data. That is the base of all the underwriting decisions, like the most fundamental layer. But on top of that, we want to be able to create models for different kind of industry and regions. For example, a
pizza restaurant in Houston, Texas, their cash flow will be very different from like a cleaning service in New York. So the differentiator, the industry is different. And second, the region is different. Like a restaurant is very different from a cleaning service. And Texas is very different from, say, New York.
So all those data needs to be contextualized in some ways. And those data are not provided by the bank account data, for example, Plaid. They only tell you the transaction. But how do we put it into context of the business, where they operate and how they operate? So by collaborating with all those AI native SaaS companies, for example, Dental Clinic SaaS, they collect data from their customers. They use AI to make their workflow better.
So by collaborating with them, we are able to get more like richer data outside of any traditional financial service companies. But all those data are related to how those businesses operate on a day-to-day basis.
And the advantage of that is that we can use AI to process those unstructured data into a more structured format. So they can use, for example, just scrape the website of the company and then from a very unstructured data to a more structured JSON, like where does it locate? What's the phone number? How's their social media operating? All those data points are not directly related in the fundamental layer, but they are all related holistically.
I think that would be the differentiator between us versus traditional players that we are able to provide a more holistic view of how the business is operating as a whole. I guess my next question is, how did you collaborate, as you mentioned, with those AI-native companies? You said they provide you with the data for specific verticals like dental clinics.
If they provided you with data, what did you provide back to them? Did you just provide your API for embedded? So currently we are still focusing on serving the SMB directly because the integration process with the partner takes time and we need to evaluate if it is a fit for both of us to invest in this integration.
So for our plan for our AI native SaaS company is that we want to be a mutually beneficial collaboration.
So the reason many AI SaaS companies want to integrate with Orbit is that they don't have all the capabilities to develop a standalone financial product that takes a lot of effort, as I mentioned before, to do the KYB, KYC, fraud detect, partnering with bank, all things like that. So they would prefer something just like a Stripe API, just like plug and play, so they can just use it and then serve the businesses of their users.
So on their side, if their users, mostly those business users have more cash on hand, they probably would spend more on their platform as well. So they probably extend their subscription or they like do some, buy some other products from the platforms.
So that makes sure that their transaction volume goes up if they integrate with Orbit. Because many times the businesses, it's not they don't want to buy it, it's simply they don't have the money to buy it. If you provide a credit that is like when they need it, they can just like access the credit. And obviously the businesses want to upgrade their equipment, want to do other stuff they need.
So that is the first. And second, for the data sharing, obviously the data will be the moat for all the underwriting decision or the financial decisions.
like tips our our partners we would like to have some sort of like a percentage give back for example full amount of transactions that they facilitate through orbit they're able to get a one percent or several percentage back so it
It's more like a distribution channel because those AI SaaS companies already reaching all those customers in niche industries, all the distribution channels out there. And then if they integrate with Orbit, they're able to serve their customer better and they're able to make more money.
So that would be like many benefits for collaborate with orbit and share data with. So I guess I, my curiosity is like, why don't they just integrate with Stripe? Like I'm very established, uh, FinTech solution instead of you guys, just because you provide more benefits.
Yes. So Stripe basically is a payment processor. They facilitate all kinds of payment. But they don't provide anything related to underwriting.
So you can strive to, say, accept credit card transaction or accept, say, PayPal or Venmo, whatever payment method you want to accept. But you cannot decide if a business is financially healthy enough to take out term loans.
So between businesses, it's very common to have net terms. So this business promised to pay the other business back in, say, 30 days, 60 days, or 90 days. And Stripe is mostly a payment process. So they don't have that capability right now.
And then for the B2B product perspective, it's also not business focused. It's more just general transaction or fintech capabilities. So what's coming next for your company? You know, there's all those being done now.
Yeah, so we're launching the beta version with a small group of local SMBs in Texas, probably next week or so.
So, my co-founder, he's a great business person and he has been working with, I would say, almost 100 different SMBs throughout his career. So, he understands the industry pretty well, especially their pain point. He has a vast network of business owners. We are...
choosing a small group of local SMBs to beta test our product. And once it reached a certain scale, we can launch to the general businesses. All right. I guess I'm curious about what background does your co-founder have? So he has been working in like different industries, like tech companies. And also he like has real estate, works in real estate. And he just like a, you
I always say a very charismatic, kind of like complete opposite of me. I think it's very complimentary in a sense that I can focus on building product, the user experience, and he understand the user needs is able to sell like crazy. Do you or your co-founder choose to build a startup in the B2B FinTech for SMB space or did he or you choose this space?
I would say he or me choose this space. We're trying to build something with AI from two years ago. We just try to make something work. We don't mind which industry would that be, but it just happened to be that the SMBs and the fintech space make the most sense to me. And because we are able to
tap into the transaction level instead of just build another subscription-based SaaS that may or may not help the business, actually. But the financing for sure would help many businesses to get out of their cashflow gaps.
Got it. Yeah, talking about AI, we've been touching a lot of topics about AI. My next question is, does AI help you to build as a founder, as a CTO? Maybe can you give some specific examples? Yes, for sure. So I would consider myself more of a CPTO, so Chief Product and Technology Officer.
I need to think about the product as a whole and also implement it. So for AI, it makes me to pick up new skills much quicker than before. So as I mentioned, I kind of did a bunch of different things in college. I didn't specialize in any of the verticals.
But AI allows me to learn stuff on demand. For example, if I need to learn front-end, then just show me how to do it. And then back-end, then just show me how to do the database, deploy, debug, all things like that.
And also, AI makes it very easy to just go from idea to implementation directly. Instead of previously, we have to design on Figma first, and then find somebody, implement it in front-end, and then find another person to implement it in back-end. So there's a lot of human labors and communication involved with our AI.
And also for many people, they may refer to it as vibe coding. They just chat with the AI agent and let it implement. But honestly, I don't do vibe coding that much. I usually don't give the authority to change my code base for an AI agent. I just ask it to think it through and then generate the code. I very intentionally choose to adopt certain code or not.
So I think vibe coding is good for like simpler project, but where code base becomes like several thousand lines of code, millions of lines of codes that become kind of hard to manage.
Yeah, that's a very interesting point that you brought up because I came across a blog article, I think, or a video about appropriate use of generative AI. And the main point it has is use AI whenever two of the following scenarios are met. Number one is you understand the area very well.
So like you already have a very good understanding about it so that you can detect if something goes wrong, if it really goes wrong. And the second one is the you are raising is you can fix the error
you can personally fix the error. So if AI messes up, messed up, you can just really quickly do the fix. Whereas in some scenarios, ask questions and you can actually detect the error it made, but actually it messed up a lot of things that you have to actually spend more time fixing the error made, the time you saved from asking it the question.
Any of the two scenarios are not met, don't use AI. And that's basically what the suggestion is. The point you brought up, it's a very good illustration. Yeah, I think it definitely for sure goes both ways. I mentioned like as a technical founder, but for non-technical founder, they can use like tools like Bolt or Replit with AI agents to build a prototype, but they would
sometimes stumble upon a problem, like an error message, but they don't have the actual coding techniques or skill sets to debug it. So they can push it to a certain level. But if you want to actually solve specific problems, then you need
at least some understanding of how things work fundamentally. There's another article that in which you wrote about you interacting with LLM, like asking you to analyze your daily logs you did for the last whole year when you finished college and then started your startup. I think that article is really interesting. Maybe can you walk through that very briefly and explain to the audience like what is your, if
initial idea that motivates you to do that and what's your main takeaway from doing that okay so actually i developed this daily journaling habit during covet through my gap year
So I was like in the quarantine, I have to find myself something to do. So I started to journal like what I actually have done for each day. And that actually ended up become, I think a great habit to help me analyze how I spend my time on a daily basis. So I think I would view my life as a, like a life language model, just like LMS predict the next token in a, in a text format or image format.
I view my life as I'm predicting my next day based on all the days I've lived in the past. So actually, I started daily journaling.
It's kind of like a process to curate my personal living data set to help me predict my next day better. So at the end of the year, for example, last year, I download all of my daily journaling data in the spreadsheet. So I download all the data and then I feed it to a local LM to analyze it. It ended up to extract all the vectors, like based on what activities I did and then cluster them into different groups.
So just by looking at that image, it becomes very clear how I spend my time for the past year. So it's kind of like a densely compressed version of my focus and my time. So I think it's very interesting just to
observe yourself or understand yourself from a reconciliation process. When you live your day to day, you don't actually think about it. But when you look at it as a whole, you started to find some interesting patterns.
So if you think, okay, I'm not being healthy as a founder, I get stressed out too much, then look at what you do on a daily basis. Do you actually work out? Do you actually meditate? Do you actually try to improve yourself by reading or learning new skills? It's just like all there in the immutable leisure of your life. So I think just a good way to stay disciplined, talking to it as a fun way to analyze it. Yeah.
What has been the biggest challenge for you as a founder? I would say just try to prioritize and focus. Just there's so many things to be done in a relatively short timeline. Like basically try to stay mentally healthy and prioritize, focus, have enough deep work each day to actually get the stuff done and then take breaks instead of like burn myself out. I think it's hard to maintain the...
Generally helps. So I highly recommend other people do it as well. Did you bootstrap or race for your current startup?
We mostly bootstrapped and we used our own money. Now let's sort of move into the last section that, you know, it's a section that we kept for all our series. It's a question for sorts of recommendations. If you would give yourself, your younger self, a piece of advice at each stage, student, a new founder, what would it be?
I think for students and founders, I think it's very important, especially in the age of AI, to just start building things and try to do things instead of try to learn everything first and then do stuff. I think it's very different from what the education system try to do.
try to do. All the high schools and colleges, usually you go to class and the professors start with a high-level principle or theory, and then eventually you start to implement something with your homework or project. So that is a very deductive reasoning kind of pattern. But with AI, sometimes
Like you just ask AI and give the result right away, but you may not understand why, but it just works. But this is a very different, like a reverse way of thinking. So you don't need to understand everything to start. You can just start and then figure out why things work along the way. Well, sometimes like AI,
like the school would say, okay, you have to follow this plan. You have to be a CS major, you know, something like that. But with the help of AI and so many tools out there, you can just start with the thing you actually want to build or the things you actually want to become, and then figure out the reason or the resources when you move along with it. I think it's a much more
efficient way, but it may not be the most like easy way. It may not fit everyone. Yeah, it may not fit everyone. But I would say if you want to do things that you've never done before or do things like achieve something more ambitious than you ever saw, you have to start with something first and then figure out things afterwards. I think that's the first. And second thing is
I think just taking small steps each day to achieve something, because small steps will compound eventually. So if you just keep doing improvement, improve yourself for 1% per day, you do it for one year, a decade, that will make a huge difference. But on a daily basis, it may be a very small notch that you barely notice. But
If you do small things consistently, that will result in a great compounding. Since most of our listeners are international students and early stage founders, do you have any particular advice for them? Yes. So as an international student myself, I find like sometimes...
I don't fit into certain criteria or certain way of thinking, but I think that is a great unique advantage for all the international students, not just in the US, but around the globe and founder as well. So I think this unique international perspective gave us a unique way of doing things and spotting the problems and giving the solutions. So
I think is a great gift and background to make use of. Actually, a lot of founders are international and they're immigrants and they have the grit to go to a new environment, start a new journey. I think it's a very, very interesting background. And for all the international students in this current era,
I think the policy and certainty, I think just making plans and then know where you're going to end up. Maybe in the short term, it wouldn't work out. But in the long term, if you keep working towards your goals, I think it will work out. I think just have patience and then don't give up. Is there any go-to source of insights, such as books, podcasts, newsletters or anything else that you would recommend to our listeners?
For AI-related topics, I recommend the daily AI newsletter. It's a newsletter curated by AI on AI topics.
So there actually are a lot of this kind of AI generated or AI curated newsletters. So you can just choose whatever topics you're interested in and the AI will scrape the internet like X or Twitter or Reddit, whatever platforms you are interested in. And you can just like summarize all the industry trends for you so you can save your time. Just stay on top of the trend for like five or 10 minutes per day.
And for FinTech related, I recommend FinTech Takes Newsletter. So it has some deep dives, FinTech topics, and some interesting blogs on there. And for general startup or market related, I usually listen to several podcasts, including the All In podcast, BG2, and Acquired.
So Acquired is relatively like deep dive, like you have to get ready for like two, three hours worth of content. But I think they did a good research on many iconic companies. I think just some good podcasts to listen to during commute. Yeah. Got it. Yeah. Thanks so much, Jian. I hope to see you again and do another channel with you after six to 12 months. Okay. Thank you so much. Thank you for your time.
Thank you. Bye.