Why We Are Building Dayos in Singapore

I spent the better part of two decades in the US tech ecosystem. I led finance engineering at Robinhood through the IPO. Before that, I spent years deep in Oracle implementations at companies across healthcare and financial services. I know Silicon Valley. I know what it offers and what it costs.

In 2023, I left all of that behind and moved to Singapore to build Dayos.

People still ask me why. The short answer is that Singapore is building something the Valley stopped building a long time ago: an entire national ecosystem designed to turn AI from a talking point into an operating system for how companies actually work. And it's doing it from the bottom up, not the top down.

The long answer is what follows. And in a world that's getting more closed and more nationalistic by the day, I think it's important to explain why an American founder chose to build elsewhere. I believe in globalization. I believe in allowing the best people to tackle the hardest problems, regardless of where they were born. Singapore believes that too.

But this isn't just a founder's story. If you're an enterprise running Oracle, SAP, Workday, or NetSuite, the place where your AI vendor builds has direct consequences for what you get, what it costs, and whether it actually works. Here's why our customers benefit from the fact that Dayos is built in Singapore.

1. The US Is Betting on Giants. Singapore Is Betting on Builders. Your Enterprise Benefits from the Builder Model.

The current American AI strategy, to the extent one exists, is to funnel capital and policy support into a handful of massive players. OpenAI. Anthropic. Google. The theory is that if you build the biggest models, everything else follows. That's a reasonable bet for foundational research. It's a terrible bet for enterprise transformation.

Singapore took the opposite approach. The government is building infrastructure, incentives, and programmes that enable small, focused companies to solve real problems inside real businesses. Over 60 companies have already established AI Centres of Excellence here. The Champions of AI programme, launching this year, is designed to find companies with the ambition to embed AI across core operations, not the biggest fundraisers.

What does this mean for you as a customer? It means Dayos operates inside an ecosystem that rewards us for delivering measurable outcomes in your environment, not for raising the biggest round or publishing the flashiest demo. Singapore's Budget 2026 introduced a 400% tax deduction for qualifying AI expenditure. The Productivity Solutions Grant is expanding to cover AI-enabled solutions for companies of any size. These incentives flow downstream: they keep our cost structure lean, so our pricing reflects the actual value delivered, not Silicon Valley overhead. 

The Career Conversion Programme reimburses employers up to 70% of salary costs for new hires transitioning into growth roles, such as AI (up to 90% for mature workers over 40). That means Dayos can invest in specialists who understand your Oracle FSM configuration or your SAP procurement workflows without passing those hiring costs through to you. Try getting that kind of support from a US state government.

In the Valley, if you're not raising a $100M round, you're invisible. In Singapore, if you're delivering outcomes, you have a seat at the table. For our customers, that focus on outcomes is the whole point.

2. The YC Model Doesn't Work for Enterprise. Our Customers Need a Different Approach.

I have enormous respect for what Y Combinator built. But the YC model has a structural blind spot: most YC companies sell to other startups. They build tools for developers, platforms for other SaaS companies, and infrastructure for the tech ecosystem. That's a fine business until you try to cross the chasm into enterprise B2B. Most never do.

Enterprise procurement, compliance requirements, integration complexity, and multi-year contract cycles. These are fundamentally different from selling a $99/month tool to a seed-stage founder. The skills, the sales motion, the product architecture, none of it transfers cleanly.

Here's what that means for you as an enterprise buyer. Most AI vendors you're evaluating were built to ship fast and iterate in 90-day accelerator windows. They optimized for demo speed, not for surviving your change management process or your security review. They've never had to reconcile 251 Outward Parcel Accounts across 140 postal authorities or automate a 4-way procurement match inside a 15-year-old Oracle EBS instance.

Singapore's AI Centre of Excellence programme provides up to 24 months of structured support with salary reimbursements that aren't capped the way an accelerator check is. That runway means Dayos can commit to 90-day outcome-based pilots where we prove value in your environment before you commit. We're not burning through a fixed pool of capital trying to find product-market fit on your dime.

Here's the example I keep coming back to: Intuit is the largest accounting software company for SMEs worldwide. Millions of small businesses run on QuickBooks. But Intuit itself? It runs its own finance operations on Oracle. The company that sells simplicity to everyone else needs enterprise-grade systems to manage its own complexity. That gap between what gets built in the Valley and what enterprises actually need to operate is enormous, and it's exactly the gap Dayos is built to close.

3. The Valley's Groupthink Problem Is Your Integration Problem.

There's a kind of architectural groupthink happening in Silicon Valley right now that should concern every enterprise buyer. Everyone is building the same way. Same model architectures. Same API-first approaches. Same assumptions about how AI agents should work.

The problem is that your world doesn't run on clean APIs and greenfield systems. It runs on Oracle EBS instances that have been customized for 15 years. It runs on SAP environments with decades of accumulated business logic. It runs on Workday configurations that reflect organizational decisions made long before anyone considered AI agents.

When a vendor builds AI that assumes the world looks like a well-documented REST API, that AI breaks the moment it touches your environment. The reason Dayos's approach works for our customers is that we start from the legacy system outward, not from the model inward. Our four reasoning strategies (Simple Feedback, ReAct, Reflection, ReWOO) are mapped to specific enterprise processes because different problems require different reasoning architectures. A journal entry automation doesn't need the same approach as a 4-way procurement match or an external reconciliation across 140 postal authorities.

That kind of thinking doesn't come from copying whatever architecture OpenAI published last quarter. It comes from spending 20 years inside these systems and understanding that true agentic AI has to do analysis and take action, not just send alerts.

Singapore's ecosystem encourages this kind of differentiated thinking because companies here are measured on deployment outcomes, not on whether their architecture matches the latest trending paper on arXiv. For our customers, that means you get AI that was purpose-built for how your systems actually work, not how a VC pitch deck wishes they worked.

4. Singapore Decided AI Is a National Priority. That Creates a Better Vendor Ecosystem for You.

There's a difference between a government saying "AI is important" and a government restructuring its entire economic strategy around it. Singapore did the latter.

Prime Minister Lawrence Wong announced a National AI Council in Budget 2026 that he personally chairs. Not a working group. Not an advisory panel. The PM, chairing it directly, with ministers from Trade and Industry, Health, Digital Development, Manpower, and Transport at the table. When's the last time a head of state put themselves in that seat for AI?

Then came the National AI Missions targeting four sectors: advanced manufacturing, connectivity, finance, and healthcare. These aren't vague aspirations. They're structured programs with curated datasets, computing resources, regulatory sandboxes, and solution providers, all designed to shorten the path from development to deployment.

And this isn't just my theory. In March 2026, Yann LeCun, one of the three Turing Award winners considered the "godfathers of AI," raised $1.03 billion for his new venture AMI Labs and named Singapore one of only four global hubs alongside Paris, New York, and Montreal. The round was backed by Singapore's Temasek and Sea, plus Nvidia and Bezos Expeditions. LeCun left Meta because he believes Silicon Valley is too fixated on the current generation of large language models. When one of the most respected minds in AI looks at every city on Earth and picks Singapore to recruit talent and be close to future clients in Asia, that tells you something about what's being built here. Singapore is attracting both foundational AI research and applied enterprise AI companies like Dayos. The full stack is forming.

For enterprise customers, this matters in practical terms. When your AI vendor operates inside a national ecosystem with dedicated compute infrastructure, regulatory clarity, and government-backed deployment programs, you get faster implementations, lower risk, and a vendor that isn't going to pivot to the consumer market because enterprise is "too hard." Deputy Prime Minister Gan Kim Yong said it plainly in Parliament: the goal is to position Singapore as a place where AI solutions are built, proven, and scaled. Not theorized about. Not demoed. Built, proven, and scaled. That language matters because it's the same standard our customers hold us to.

5. Regulation That Works for Enterprise Buyers, Not Against Them.

In the US, AI regulation is a political football. In Singapore, it's a competitive advantage, and that advantage flows directly to our customers.

The Monetary Authority of Singapore published AI Risk Management Guidelines covering governance, lifecycle controls, third-party vendor risk, fairness, explainability, and human oversight. These aren't punitive rules designed to slow things down. They're frameworks that give financial institutions the confidence to actually deploy AI, because they know the guardrails exist.

Dayos is SOC 2 Type 2 audited and ISO 42001 certified. In Silicon Valley, people's eyes glaze over when you mention compliance certifications. In Singapore and across regulated industries in APAC, those certifications open doors. They signal that we're serious about building AI that works inside your compliance framework, not just AI that demos well on a stage.

For customers in financial services, healthcare, and logistics, this is not abstract. Singapore's approach is proportionate and risk-based. That means Dayos isn't buried under the same requirements as DBS or OCBC, but we're held to a standard that earns your trust. And that trust is what gets you from pilot to production. If your vendor can't pass a SOC 2 audit or articulate their AI governance framework, how confident should you be putting them inside your financial close process?

6. Quality of Life Produces Better Work for Customers.

There's a practical dimension to choosing where you build a company, and it has downstream effects on the quality of what customers receive.

Building a startup demands everything you have. It demands it for years. If you're also a parent, the environment around you determines whether you can actually sustain that. Singapore has a support infrastructure, from domestic helpers to accessible childcare, that enables parents to create and build at the same level as a 22-year-old with no responsibilities. I've watched talented founders in the US step back from companies not because they ran out of ideas, but because the system around them made it impossible to be both a present parent and a committed founder.

And I don't think about school safety in Singapore. I don't worry about gun violence. That cognitive load is real, and its absence frees up something that's hard to quantify but impossible to ignore.

For customers, the implication is straightforward: a founding team that isn't burning out, isn't distracted by the cost of living crisis, and isn't operating in survival mode produces better product, better support, and more consistent delivery. The sustainability of your vendor matters as much as their technology.

7. The Talent Is Already Here. And We're Building the Next Generation.

Here's something most enterprise buyers don't think about when evaluating AI vendors: where does the vendor's engineering talent actually come from, and how close are they to the people who already support your systems?

The answer, for most US-based enterprises, might surprise you. The experienced systems engineers who maintain your Oracle, SAP, and Workday environments, the people who know your configurations, your customizations, your business logic, the vast majority of them live across APAC. Two decades of global outsourcing moved the center of gravity for enterprise application support to India, the Philippines, Malaysia, and Singapore. The talent pool that understands your ERP inside and out is already in our time zone and our region.

By building in Singapore, Dayos sits at the center of that talent ecosystem. We're not trying to recruit Oracle specialists in San Francisco where they barely exist and cost $250K+. We're hiring from the region where this expertise is concentrated, at cost structures that let us invest more in your outcomes and less in Bay Area rent.

But we're also building for the future. Dayos has established active partnerships with the career centers and libraries at NUS, SMU, and NTU, Singapore's three leading universities. These partnerships give us direct access to graduating software engineers, data scientists, and business technology students who are trained in one of the world's most rigorous education systems. We sponsor projects, mentor students, host workshops, and create a pipeline of talent that understands enterprise systems and AI from day one of their careers.

Singapore's universities are globally ranked and intensely competitive. NUS and NTU consistently place in the top 15 worldwide for computer science and engineering. SMU produces some of the strongest business technology graduates in Asia. The graduates coming out of these programs aren't just technically skilled, they're trained to operate in a multicultural, globally connected business environment, which is exactly what enterprise AI deployment demands.

For our customers, this means three things. First, Dayos has access to senior systems engineers who already understand your technology stack because they've been supporting companies like yours for years. Second, we're building a next-generation talent pipeline through Singapore's world-class universities, which means we can scale without sacrificing quality. Third, our blended team of experienced practitioners and emerging talent keeps our delivery model both deep and cost-effective, and those savings get passed through to you.

The Honest Comparison

Silicon Valley still has more capital, more brand recognition, and a deeper bench of serial entrepreneurs than anywhere else in the world. I'm not pretending otherwise.

Here's what Singapore has that the Valley doesn't, and what it means for enterprise customers who choose to work with a vendor built here:

A government that treats AI deployment as a national economic strategy, not a talking point. An ecosystem that supports focused AI companies instead of only rewarding scale. Regulatory frameworks that enable enterprise adoption instead of paralyzing it. A physical ecosystem being purpose-built for AI companies at Kampong AI. Tax incentives that directly reduce the cost of building and deploying. English as the official language of business and government, which no other country in Asia has. A geographic position at the center of ASEAN, the world's fourth-largest economic bloc by 2030. Proximity to the talent pool that already supports most global enterprise systems. And a culture that values outcomes over hype.

When I was at Robinhood, I watched a company go from startup to IPO. I know what hypergrowth looks like in the US. But I also know that enterprise AI needs something different. It needs patience, trust, regulatory alignment, and a market that rewards companies for actually delivering results.

Singapore is building exactly that market. And Dayos is building right alongside it.