Job at a glance
Who is DexCare? DexCare optimizes time in healthcare, streamlining patient access, reducing waits, and enhancing overall experiences. Guided by our mission – Nobody waits for care – DexCare addresses tech gaps by aligning supply and demand to modernize healthcare infrastructures for an inclusive ecosystem. We are equally committed to our DEIB mission: Creating an inclusive workplace where diversity drives innovation, equity ensures fairness, and belonging strengthens collaboration, enabling everyone to thrive. What is DexCare?
DexCare, a digital care orchestration platform, streamlines care delivery logistics. It empowers healthcare systems to predict constraints and precisely schedule services, optimizecapacity, and cuts operational costs. Currently serving 57 million patients, including Kaiser Permanente and Providence, DexCare ushers in a new era of digital-care access, ensuring health systems can efficiently track and deliver every hour of capacity for consumer ease.
For more information, visit www.dexcare.com or follow us on LinkedIn . About the role AI is now central to how patients reach care at DexCare. Conversational agents talk to patients by voice and text. Predictive models decide where capacity should go. Knowledge systems turn a health system's undocumented operational rules into something machines can act on. All of it is live with US customers today.
We're hiring a VP of AI/ML to own that work end to end, reporting to the CTO. You'll set technical direction for AI/ML across the company, build and scale the ML engineering team in US / India, and be accountable for how these systems behave in production — where a wrong answer to a patient is a safety event, not a bad demo. This is a senior leadership role for an engineer at heart. You'll build and manage a team and still be expected to review architecture, read code, and hold a real opinion about model behavior.
You'll work closely with engineering, product, and data leaders based in the US. What you'll own Conversational AI. The engine behind patient-facing voice and text agents — orchestration, dialogue design, and the balance between model-driven flexibility and the deterministic control healthcare demands. Realtime voice. Latency, concurrency, and conversation quality on live telephony at production volume.
Predictive ML. Demand, capacity, ranking and matching problems across the scheduling and access surface. Retrieval and knowledge. Extracting structure from clinical and operational material that was never meant to be machine-readable, and serving it back to agents and applications reliably. Safety and reliability. Classification and monitoring that runs alongside every patient conversation, plus the evaluation systems that decide whether a change is safe to ship.
The full lifecycle. Experimentation through deployment, monitoring, and the MLOps practice underneath it. Model strategy and economics. Build-versus-buy across model providers, and the unit cost of every interaction as a metric you manage deliberately. The team. Hiring, structuring and growing the ML organization in US / India as part of a global function. What you'll bring You've shipped LLM-based systems into production under real regulatory constraint — healthcare, financial services, or somewhere else a wrong answer has consequences.
Beyond that: Typically 15+ years across machine learning, AI, software engineering or data science, with a hands-on engineering foundation Deep Python and modern ML tooling; PyTorch, TensorFlow or equivalent Realtime or streaming voice systems: latency budgets, interruption handling, telephony Retrieval systems built over messy enterprise knowledge Evaluation harnesses that gate releases Experience hiring and leading ML engineering teams and senior technical talent, ideally including building a team from a small base Multi-tenant SaaS, with a working understanding of tenant isolation and per-customer configuration Production experien