There is a particular kind of clarity that comes from building a life far away from where you started.
When you grow up somewhere, you do not see the systems around you. The healthcare infrastructure, the institutional scaffolding, the invisible support structures of daily life become invisible through familiarity. You only begin to see what is absent when you leave, and then return, or carry the memory of home with you long enough to measure it honestly against what you now know is possible.
Mohieminul Khan and I both came to the United States from Bangladesh as young students. We built careers here, in technology. We learned how software systems are designed, how artificial intelligence works, and what becomes possible when engineering skill and product thinking are concentrated on a real problem. In the way that immigrants often do, we became fluent in two worlds simultaneously.
We eventually arrived at the same question, independently and then together: why is none of this being built for home?
What Distance Teaches You
Growing up in Bangladesh, the limitations of the healthcare system were not statistics. They were your grandmother waiting hours to be seen. They were the pharmacist who also served as a medical advisor because that was who was available. They were clinical guidelines calibrated for conditions and resources that bore little resemblance to the actual circumstances of people's lives.
When you spend years working in American technology, you see what becomes possible when talent and resources are directed at a problem in earnest. And you cannot stop thinking about the problems that have never received that kind of attention.
The maternal health gap in Bangladesh is one of those problems. Four million births per year. Postpartum depression rates among the highest documented in the world. No AI tools built specifically for Bengali-speaking mothers, in their language, for their context. The gap is vast and, once you see it clearly, impossible to rationalize.
Why Maternal Health, and Why First
Identifying a gap is not the same as knowing where to intervene. We spent considerable time asking a direct question: where does a technology intervention have the most potential to produce a meaningful and lasting difference?
Maternal and infant health kept surfacing as the answer.
The first two years of life, including the prenatal and postpartum period, are universally recognized by the World Health Organization, UNICEF, and the global health research community as the period of maximum developmental consequence. What happens during this window shapes a person's health trajectory, cognitive development, and life outcomes in ways that are genuinely difficult to reverse at later stages. The science on this point is consistent and well-established.
For Bangladeshi mothers, this critical window is typically navigated without adequate information or support. Traditional knowledge is valuable but incomplete. Community health workers are stretched far beyond their capacity. Accurate, accessible, culturally grounded health information in Bengali is genuinely scarce.
An AI companion that could support mothers between clinic visits, at odd hours, and during moments of uncertainty seemed like exactly the right starting point.
What Building for a Community You Care About Actually Feels Like
We want to be direct about something that is not often discussed in product narratives: building for a community you care about is harder in specific ways than building for an abstract market.
Every decision carries weight. Is this feature actually useful, or are we building what we imagine a Bangladeshi mother needs rather than what she says she needs? Is the language natural, or does it read like a government document translated by someone who is technically correct but culturally distant? Are we listening, or are we projecting?
We talked with mothers. We talked with community health workers in Bangladesh. We talked with doctors. We read extensively on maternal health in South Asia, on Bengali NLP challenges, on how health information is actually transmitted and trusted in Bangladeshi communities.
The product that resulted from that process is not what we originally imagined. It is better, because it was shaped by the people it exists to serve, not only by us.
The Name We Chose for the AI
One detail that we think matters: the AI companion inside Hafsa Sastho is named Hafsa Apa. Apa is the Bengali word for elder sister. It is a term of genuine affection, trust, and reliability.
We did not name her an assistant, a bot, or a system. We named her elder sister because that is the relationship the product is meant to embody. The person you call at any hour with a question you feel embarrassed to ask elsewhere. The person who gives you real information rather than only reassurance. The person who takes your experience seriously and responds without judgment.
That is the standard we are trying to meet. Not a product interaction. A relationship.
What We Believe About Technology and Equity
Our perspective on this is direct: the first wave of AI innovation produced remarkable value, and it distributed that value unequally. The tools and platforms that defined the past decade were built primarily for populations that were already well-served by existing digital infrastructure.
We believe the next wave has to be different, not primarily as an ethical aspiration but as a practical commitment. The problems that remain unsolved affect enormous numbers of people and represent genuine opportunities to build things that matter.
Building for underserved communities requires the same rigor, the same discipline, and the same relentless focus on actually working in the real world as building for any other market. In many respects, it requires more. The margin for products that do not actually work is smaller. The trust that must be earned is harder-won.
We came from Bangladesh. We came to America. We learned what we could. And now we are trying to bring what we have learned back to the people and places that could benefit from it most.
That is the whole story.
Nahl Technologies is based in Indiana, USA. For partnership inquiries, investor conversations, or to apply for the Hafsa Sastho beta program, visit nahltech.com/contact.