Jerry Shu, Daylit Co-Founder & CTO on AI for Enterprise Finance

**Grace Gong** (0:02)
Oh, we're live. Hi Jerry, welcome to Venture with Grace.

**Jerry Shu** (0:06)
Hey Grace, yeah, nice to be here.

**Grace Gong** (0:09)
Before we start off our show, I want to give a quick shout out to our amazing sponsor. This episode is brought to you by Nebius, the ultimate cloud for AI novayers. Nebius provides AI infrastructure you can count on, combining reliability and speed with flexibility and engineering support unmatched by hyperscalers. AI leaders like Meta, Shopify, Higgsfield, are already partnered with Nebius to run their AI workloads. Plus, venture-backed startup can save up to 150k on compute cost with all cloud for access. Visit nebius.com or nebius.com/startup to learn more.
Okay, I want to jump into, so I want to start with your background. So you have been, so I feel like you have been a trader, no?

**Jerry Shu** (0:51)
Yeah, I was doing quant trading at JP. Morgan for quite a bit.

**Grace Gong** (0:57)
Yeah, and then you co-founded a different company. Maybe we could talk about what were some core lessons that you've learned early on or kind of shaped into who you are today.

**Jerry Shu** (1:07)
Yeah, definitely. I think, I mean, it's a big topic. I think a big part is really knowing what gets you excited, what you like doing, what's your real strength.
And that, I think, in general helps me a lot. I kind of figuring out what to do, what I'm really good at, and eventually kind of leads to where I am today, kind of building on Deloitte with Jared, my co-founder, kind of is a very exciting journey. Personally, I think kind of dive in a little bit. I think in particular is my background. I think I really love, from technical side, like all of these from time series prediction to all these latest back then machine learning, now AI techniques, from reinforcement learning to all of these now transformers, and then really how to use these techniques to build something that's really valuable for our day-to-day life. I think that's really exciting to me. So, early on, I really learned what really excites me is really, yeah, building innovative solution, leveraging the latest techniques that can really advance us forward. I think that's genuinely, I'm just super excited about that.
That makes me happy and I want to spend every second of my life doing. That's probably my biggest lesson. Being on the coin trading side, there's a huge, like from the dollar impact perspective, there's a lot of sharpness, kind of trained, I guess myself, understanding data, trying to understand the markets, trying to understand what matters to public, to the markets, like financial markets in general. I think these are all later becomes kind of skillsets, kind of underfolded in this direction, where I want to build something cool, and build something useful. So that's really what brings me where I am today.

**Grace Gong** (3:19)
For sure. So Daylit is basically, essentially you guys help a company to charge better. So I personally feel like I, as a small company myself, I actually feel like this is such a useful thing, because of people are still on their resume, about how long it actually takes for people to manually do it. Maybe we could start from the product itself. So obviously, I actually made a notebook out of Daylit before I come on. So essentially what you guys do is, in the past, when people are charging, let's call it a $1 million project or something, there's different milestones, and the core difference of the agentic version versus the pre-VR version is you guys really have the agent response to a very unique situation with more context, instead of just spamming people with like, hey, pay me, pay me, every two months.
So I guess maybe we could let you talk about the product first, how you guys came up with the product, and we can dive into everything technical.

**Jerry Shu** (4:22)
Yeah, definitely. So I think the product, one sentence in that show is really this AI agents for B2B accounts receivable. Then as Grace, you said nicely, the main pain point is it's one hand to work hard and perform your service. It's completely a different story to getting paid.
That process is paying for a layer of reasons, from identifying the willingness to pay to ability to pay, navigating the difference between, hey, I want to maintain customer relationship, but I also want to get paid, all the way to how can I make that process smoother, trying to understand why things gets blocked in the middle.
It's a big chunk of that, and we really try to be expert of this, and build this agentic flow to try to solve this problem. I think from Jerry and my perspective, it's really trying to remove this notion of mundane from people's daily life. That's where we think we are bringing value to folks.