Top Decile Podcast: Vijay Rajendran and Amit Goel

In 2026, nearly every firm claims to invest in AI. gAI Ventures Fund is chasing something narrower: the domain expert. On the Top Decile Podcast, co-founders Vijay Rajendran and Amit Goel sat down with Connor Sattely of Decile Group to make the case for a firm built on one contrarian idea, vertical first and AI second. Rajendran led portfolio value at 500 Global, a portfolio of roughly 2,800 companies across 80 countries. Goel is an exited operator with 20 years inside payments and identity. Together they've built a US pre-seed venture builder that acts as an institutional co-founder, not a passive check.

Watch The Full Episode Here:

The order matters, and Rajendran argues it's the right one. "It's important for somebody to be able to see the value in adopting anything right away," he said. "Otherwise, it just becomes a toy. It becomes something that's a novelty, but not a business necessity." He points to widely cited research suggesting most enterprise AI deployments have failed to deliver, and reads the takeaway plainly: everybody is talking about AI, but very few people are seeing the benefits.

Goel frames the same idea from the operator's side. "A task is not a job," he said, using radiology as the example. Deep learning can read a scan, but the job still includes boards, sign-off, liability, and a dozen other processes a model can't own. "90% to 95% job can be done by AI," he said. "The remaining is the real world, where it includes regulatory aspects, compliance aspects, and there are so many other things that you need that human in the loop." Solve that last mile with a domain expert, and the vertical AI company gets much easier to build.

gAI isn't writing checks and waiting. It builds. Rajendran ties the model to a broader shift: "The cost of human cognition is really collapsing." That collapse makes it practical to build alongside founders in a way that would have been inefficient a few years ago. Goel says the technology itself demanded a hands-on approach. gAI operates almost like an R&D function, experimenting with models and building agentic systems in-house, led by a CTO who was an AI researcher at a lab. "We actually help these founders almost build most of the initial first two versions of the product and do all the engineering work from an AI perspective," he said.

The fund is deliberately scoped. "Our thesis covers financial services, commerce, enterprise productivity," Rajendran said, all fields where the founders have built and invested before. Two portfolio companies show the pattern. The first is FastTracker AI, which Goel described as "focused on the RIA space, wealth management space," building AI-native tooling around the software advisors already use. Its founder has about 16 years in financial services and wealth management. The second, Swake AI, is "focused on small business lending space, starting with AI tooling for invoice factoring." Its founder worked in lending at JP Morgan Chase and Capital One, then at a lending fintech that went public. In both cases the founder is a top-percentile expert in a hyper-verticalized area, exactly the profile gAI is built to back.

gAI calls its build phase the "minus one to one journey." It starts before there's even a second person in the garage, moving through product, engineering, and go-to-market. Companies get carved out once they've reached roughly 15 to 20 B2B customers, cleared basics like SOC 2 Type 2, and stabilized the product. The equity structure is where Goel says they've innovated. gAI gives "majority of the equity to the founder," so that even at Series A the founder holds substantial ownership and stays motivated to build for five to ten years. His advice to anyone starting a venture builder is to treat the model itself as a product: "The structure itself, the framework itself should be pretty much like a product," set early so the team can spend its energy on the market instead.

Rajendran describes gAI's backers as an "operator coalition," successful executives and former founders whose first question is usually how they can get involved. That network sources talent, sharpens market judgment, and opens doors for portfolio companies. gAI also counts Decile as its first institutional backer, one of five start funds Decile supported last year. For LPs weighing the model, Rajendran's own hard-won lesson is concreteness over promises: "Everybody can have a thesis, anyone can make a promise," but showing the actual companies in the pipeline is what earns conviction. For domain-expert founders, the offer is simple: gAI brings the AI and engineering, the founder brings the network and expertise, and the equity is weighted to the person doing the building.

Learn more about the fund and request access on Decile Access: https://decileaccess.com/funds/gAI-Ventures-Fund

gAI Ventures Fund is a US pre-seed firm backing vertical AI startups in financial services, enterprise productivity, and commerce. Rather than writing passive checks, gAI acts as an institutional co-founder, building companies from scratch alongside domain experts. The fund is co-founded by Vijay Rajendran, who led portfolio value at 500 Global, and Amit Goel, an exited operator with two decades in payments and identity.

Watch The Full Episode Here: https://www.youtube.com/watch?v=TT2WdlXNPHY