This is my day zero log of my journey to build a new system for venture, from the initial thought of joining a venture capital firm as an Entrepreneur-in-Residence (EIR) to the process of designing the architecture of the new era.
As a former founder, I've seen how VCs operate from various angles. I was fortunate to work on several projects that allowed me to gain experience and inspired the concept of an AI-Native VC.
Looking back, it’s hard to believe that it’s already been nine months since the idea of joining the VC world first came to mind, all the way to finally having my own fund.
Here, I’ll share in detail the challenges I faced, where I gathered information, the advice I received, and how I put everything into action. I’ll share as much as I can. For the parts I can’t talk about, well… you’ll just have to ask me in person!
Should I start a fund?
When we do anything, whether starting a business or launching a new project, it all comes back to one fundamental question: Why do you want to do this?
I still remember having a conversation with a friend once. She’s a successful entrepreneur whose company was listed on Nasdaq. She usually keeps a close eye on the public markets and startup trends.
She asked me, “Why did you choose venture capital over investment banking, or even starting your own company again?”
That was such a great question.
She wasn’t the first to ask me this after I began my career in VC. Friends who know me well have often said I’d make an excellent founder, whether because of my combined engineering and business background, or my “superpower” of being able to communicate with people from all kinds of fields.
I remember my reply at the time: “I enjoy working with the best people across different industries. If I were running my own startup, I’d lose that joy.”
But looking back, I realized my answer didn’t fully explain the reasoning behind my choice.
So here, I want to answer this question from three different angles.
Vision
I remember when my last startup had just begun, a wise partner at a VC once asked me a question. At the time, I didn’t understand why he asked it, nor did I know how to answer. It wasn’t until that company had come to an end, and I was preparing to take a career break to explore new directions, that I finally understood what he meant.
He asked me: “What do you really want?”
It’s a question that can be very simple, yet also incredibly difficult to answer. You can respond on a surface level, or you can answer in a way that reflects the length and depth of your entire life.
This question shapes every decision and judgment you make. It affects how much commitment you put into a relationship—whether it’s a friendship, a romantic relationship, or a business partnership.
Luckily, I know exactly what my vision and mission are at this moment.
I’ve heard more than once from books, articles, and podcasts that running a fund is essentially a long-term relationship. Even if the capital is deployed within 2–3 years, it can take a decade for the portfolio to become liquid, all while providing continuous support to the companies you’ve invested in.
So, what is your motivation to start, and what keeps you resilient enough to push through, whether it's reaching a milestone or pursuing your ultimate north star?
I’ll never forget the moment when the idea of “supporting startups” just popped into my head at a time when I had no clear direction. It perfectly aligned with my mission and, in my view, was the most effective way to achieve my vision. Plus, it allows me to maintain an endless sense of curiosity every time I talk with brilliant founders.
I’m a generalist, and my personal interests give me the ability to have engaging conversations with top minds from all kinds of fields. Not just technology, but also music, sports, science, and beyond. Many friends think my work must be exhausting because I constantly have to absorb knowledge from different areas. But I find it incredibly fun. For me, it’s like a way to constantly fire up my brain’s neurons — something I’ve loved since I was a kid. At least it’s a pretty healthy habit (oops).
Capability
Thanks to this article, I now understand that a track record can take many forms, but it must prove that I’m the right person to manage the fund.
Deal flow is probably the easiest and most solid skill I have. I have a strong developer network across different domains and regions, built from my professional experience, global programs, nonprofit hackathons, and roles as a mentor or judge.
People tend to reach out to me both when they’re just starting to have an idea and when they’re at the final stage, still hesitating about whether to start a company.
I once asked them why they always come to me. They said it’s because of my dual background, which allows me to understand the complexity of technology, the challenges of bringing a product to market, and the current level of market acceptance.
My personality also pushes me to give the most direct analysis and feedback, which helps them quickly decide how to adjust their ideas, balance the scope and goals, and plan hiring.
I remember listening to Naval’s podcast and reading books by Peter Thiel and Jason Calacanis. I deeply resonated with their point that judgment is the hardest skill to master. There are so many approaches, techniques, and personal experiences that shape your judgment.
As someone who has loved math since childhood, I used to think we could create models for “judgment.” But once I started seeing real deals and evaluating whether to invest, I realized it’s not that simple.
Of course, I still build certain mathematical models into my agentic system, but you don’t want the decision-making process to rely entirely on models—whether traditional math models or ML/GenAI models.
Sometimes you have to consider overfitting, but more often, it’s about market sentiment. The tricky part is, in public markets, sentiment can be understood by analyzing various information flows. But at the super early stage, how do you predict market sentiment five or even ten years into the future?
That’s an art beyond math.
Timing
The VC world is more competitive than ever. Some startups today can become profitable without ever taking money from VCs. On top of that, big-name VCs have long-term brand power that gives them better access to deals. For a new VC, before you even have your own unique value proposition, even if you manage to spot an outstanding team and write them the first check, how will you actually gather the right resources to help them set sail?
I’ve read a lot of insights on this: in the past, top VCs were like customers at a conveyor-belt sushi bar—deals kept coming by non-stop, and they just had to pick the ones they liked. But does this approach still work today? If it does, why are so many large VCs now offering early, hands-on “services” to founders?
This question puzzled me for a long time. I remember chatting with a very successful Silicon Valley investor—someone who founded a company during the early internet era, took it public, and later sold it. From his perspective, strategy and resources should be something founders figure out themselves. His role, as he saw it, was simply to invest in exceptional people. But if investors need to “help” founders build their companies, what does that say about the founding team’s composition? And what exactly is the role of the investor then?
I consider myself lucky to have started building AI products back in 2017, before the general public even realized what AI could do. As the VC market was just starting to heat up, I was preparing to enter it, planning to leverage my previous skills to create a new kind of VC.
Now is the perfect time. Experience in SaaS is still valuable, but AI is not just another technology or business model shift. It’s a complete transformation in how information flows—changing speed, pipelines, and even user behavior in massive ways.
Founders will need a new kind of venture capital to adapt to this wave.