Observing the frontier, while building for our local reality

As an AI startup founder in Southeast Asia, I have spent the past year concentrating on three key aspects: achieving commercial traction, securing capital, and assembling a robust team. However, a surprising development emerged in the course of just a few months: agentic AI turned out to be a game-changer for all three areas simultaneously.

While the traditional elements of success like product, pricing, and execution remain crucial, a market's capacity to learn and adapt has emerged as a comparable essential factor. This could be seen as a reflection of timing. Some key indicators I consider include whether organizations are redesigning workflows instead of merely automating existing processes, whether leaders allocate budget for experimentation before ROI is proven, and whether customer inquiries center around implementation rather than just understanding what AI entails.

Notably, in several meetings across Asia, we were asked how we differentiated from Claude, specifically, as opposed to other retail platforms or AI applications. This question has not been posed in the US. This observation suggests the stage of AI adoption in the market – when the entire AI stack is perceived as one conversation by customers, it indicates an earlier stage in distinguishing between infrastructure, model, and application layers.

In markets that are further along, the focus shifts towards orchestration, governance, evaluation, and workflow redesign, moving beyond the fundamental question of "what is AI?" to "how do we operationalize it?" Although this insight is derived from limited conversations, it has significantly influenced my approach to go-to-market sequencing.

Capital tends to flow towards markets where customers are already experimenting, purchasing, and proving new categories. This observation explains why US AI companies continue to attract substantial investments, not just due to the depth of capital markets, but also because of the density of customers willing to test and scale frontier technology early. Venture capitalists in the US, being curious and on the same learning journey as us founders, are also advising founders to consider moving towards the US.

However, for Southeast Asia, the goal isn't to replicate Silicon Valley's capital markets, but rather to cultivate customer-adoption density that attracts capital. There is still a long way to go in terms of building the skills bar for engineers. While leaner technology teams can achieve more than ever before, the lack of a playbook for creating these teams is surprising.

The shift is moving away from traditional role-based hiring to a focus on learning velocity. Technical ability remains important, but an engineer's ability to quickly adapt to new tools and frameworks has become equally crucial, given the rapid turnover in models, orchestration patterns, and evaluation techniques.

Our engineering team, based entirely in Southeast Asia, has had to confront the engineering labor market's readiness for AI adoption. This experience has led us to prioritize curiosity and adaptability over years of experience with specific toolsets during the hiring process. Building a strong pool of technical advisors to mentor these engineers has become a top priority in our hiring strategy. This approach is based on our own team's experience and not a universal hiring formula.

Recognizing the capacity of an organization or market to learn and adapt has become increasingly important as a founder. Continuous education has become a pivotal role for us, guiding our strategies in terms of traction, fundraising, and hiring. The evolving landscape of AI adoption has significantly shaped my perspective on these critical aspects.