Software Engineer

Senior Software Engineer H2U HealthAbout H2U

H2U is Taiwan's leading digital health company, building the infrastructure for how people understand and act on their own health. Our ecosystem spans H2U pano (a unified health platform with 3 million members), H2U eXamine (Taiwan's #1 health checkup system, serving 2 million people annually), H2U eXpert (enterprise health services for 500+ companies), and Everyday Health (Taiwan's largest health media, 10M+ monthly page views).

We're not a single product we're a platform integrating health checkup data, behavioral data, medical content, and AI-driven recommendations into a coherent experience. The engineering challenge is proportional to that ambition.

The Role

We're hiring a Senior Software Engineer to join a small, high-ownership engineering team at a critical moment: we're consolidating five acquired brands into one unified platform, building AI capabilities from the ground up, and preparing for IPO and international expansion.

This is not a role where you execute tickets in a well-defined system. The architecture is still being shaped, the data landscape is complex, and the decisions you make in the next 12 months will define how H2U's technical foundation scales. We're looking for engineers who are energized by that kind of environment not intimidated by it.

Depending on your background and interests, you'll be matched to one of three focus areas after the interview process. All three work closely together.

Focus AreasProduct Engineering

You own the user-facing experience across H2U pano, Everyday Health, and eXamine health journey flows, subscription experiences, AI feature interfaces, and health data visualizations that 3 million members interact with. Your north star is whether users can actually understand and act on their health data, not just whether the feature shipped.

Platform & Infrastructure Engineering

You own the foundation that everything else runs on. Three concrete problems define this work right now: establishing observability across five brand systems so we stop finding out about incidents from user reports; building the shared service layer data, notification, membership, content APIs so the rest of the team can build without touching five different legacy stacks; and ensuring the security and compliance posture of a platform handling 2 million health checkup records per year.

AI & Data Engineering

You own the AI product loop from raw health data to meaningful user output. H2U has an asset most AI teams would envy: structured checkup data from 2 million annual health screenings, a content library of millions of health articles, and behavioral data across diet, exercise, and health events. Your job is to turn that into working AI products: RAG pipelines for health report interpretation, personalized recommendation systems, automated data labeling, and the evaluation infrastructure to know when the outputs are actually good.

What We're Looking For

We don't sort candidates by specialty at the application stage. Across all focus areas, we hire for the same core qualities.

You make decisions under ambiguity. You don't wait for complete information to start thinking. You form hypotheses, move forward, and adjust when you learn something new. At H2U, the spec is often incomplete, the data is often messy, and the right answer is rarely obvious that's the job.

You think in systems, not features. You understand that what you build has upstream and downstream effects. A data schema decision made today affects what the AI team can do in six months. An API contract chosen without thought becomes someone else's constraint for years. You factor this in without being paralyzed by it.

Your definition of done includes what happens after launch. You look at the data after a feature ships. You notice when something isn't working the way you expected. You're the first one to raise the flag not the last.

You can translate technical risk into human language. You can explain to a PM or executive why a seemingly small architectural decision matters without hiding behind jargon, and without waiting until the problem is already on fire.

Technical Expectations

We don't hire for specific frameworks. We hire for depth of thinking and a track record of execution. Regardless of focus area, 5+ years of engineering experience is expected, with clear evidence of independent ownership not just contribution.

• Product Engineering You're comfortable moving across the full stack TypeScript / React / Next.js on the frontend, Node.js or Python on the backend, REST or GraphQL APIs in between. React Native experience is a plus. What we actually evaluate is product judgment, component architecture, and whether you've owned a feature end-to-end in a real product company from zero to shipped to iterated on.
• Platform & Infrastructure Engineering You work in Python/Java/C#, manage infrastructure with Terraform or equivalent IaC, and have hands-on experience with Kubernetes and container orchestration. You understand OAuth 2.0 / OIDC well enough to design an auth system, not just integrate one. What we evaluate is whether you've built observability or platform architecture that had real operational impact not just set up tools someone else designed.
• AI & Data Engineering You work primarily in Python, are fluent in SQL, and have direct experience with vector databases (Pinecone, FAISS, Qdrant, or equivalent). What we evaluate is whether you've shipped a RAG pipeline in production and built an evaluation framework that goes beyond eyeballing outputs because in a health context, it seems fine is not a standard we can accept.

What This Role Is Not

• Not a role where you implement specs handed down from above we expect you to push back, propose alternatives, and co-own decisions
• Not a role where the system works is enough in a health platform, reliability and output accuracy are user trust, not just engineering metrics
• Not a research role everything we build is measured against real user behavior and health outcomes

Engineering Philosophy

• Trust is the product. H2U handles some of the most sensitive data a person has their health. Every engineering decision, whether it's an API design, a model output threshold, or an alert configuration, either builds or erodes that trust. We take that seriously.
• Simplicity is a strategy, not a compromise. We're integrating five legacy systems while building new capabilities simultaneously. In that context, every layer of unnecessary complexity has a real operational cost. We default to the simplest thing that works and add complexity only when we can articulate exactly why.
• Velocity with accountability. We move fast because the market and the mission require it. But we ship things we're willing to stand behind and we instrument everything so we know quickly when we're wrong.