Building AI Solutions to Transform Health Care

By the end of the program, participants will be able to: 

• Identify and prioritize high-value health care problems suited for AI-enabled intervention
• Translate complex workflow or service problems into clear product opportunities
• Use structured AI-native methods to pressure-test assumptions across clinical, operational, technical, safety, and policy dimensions
• Develop a Product Requirements Document (PRD) including users, workflows, success metrics, and acceptance criteria
• Build and test a functional prototype or substantial MVP
• Evaluate implementation readiness, governance needs, and pilot design considerations
• Communicate the opportunity, product, and next-step plan through a concise decision-ready pitch

Please note that this information is subject to change.

Four live online sessions delivered over two weeks. Days 1 and 2 focus on selecting and validating a high-value opportunity. Between weeks, participants complete a structured sprint to refine their concept and prepare for the build sessions. Days 3 and 4 focus on product specification, prototyping, and the pathway from prototype to pilot.

Daily format

• 9:00–10:00 AM ET: Faculty session
• 10:00–10:30 AM ET: Live Q&A
• 11:00 AM–12:00 PM ET: Guided lab
• 12:00–12:30 PM ET: Debrief and office hours
• 12:30–1:30 PM ET: Small-group build studio
• 1:30–2:00 PM ET: Faculty feedback and next steps

NB On the last day the final session runs for one hour from 1:30-2:30.

Please note that this information is subject to change.

Day 1: Identify and frame the opportunity

Objective: Understand where AI can create measurable value in health systems and select a problem that is important, specific, and buildable.

Faculty session
AI for Health Systems Impact (Atun)
Participants examine the global health systems context and the incentives shaping providers, payers, governments, life sciences organizations, and health platforms. The session then turns to what generative AI and agentic systems change in practice, and how leaders can identify process, workflow, and service failures that are suitable for AI-enabled redesign.

Guided lab
Rapid Prototyping with AI Tools (Panch)
Participants are introduced to AI-assisted product development, including the fundamentals of software building for non-engineers, what rapid prototyping can and cannot do, and how to translate a problem statement into a first-pass solution concept. The emphasis is on disciplined prototyping rather than ad hoc prompting.

Build studio
With support from technical faculty, participants source and scope a high-leverage problem, define the target user, map the current-state workflow, articulate the pain point, and generate initial product concepts in the Healthcare AI Studio. They then critique those concepts and identify key limitations, assumptions, and open questions.

Outputs
Opportunity brief; target user and current-state workflow; initial solution concepts; preferred concept with key limitations and open questions.

Day 2: Validate the use case and define the MVP

Objective
Test the selected opportunity across stakeholder, technical, clinical, operational, safety, and regulatory dimensions, and define a realistic MVP.

Faculty session
Developing Health Care Innovations and Aligning Stakeholder Value (Atun)
This session focuses on how value is created and captured across clinicians, operators, patients, payers, regulators, and organizational leadership. Participants learn how to assess whether an idea is compelling, feasible, safe, and worth building.

Guided lab
Structured Validation in the Healthcare AI Studio (Technical faculty)
Participants use the Studio and AI expert panels to pressure-test their concept from clinical, operational, regulatory, technical, and implementation perspectives. The goal is to move from a promising idea to a validated use case with clear boundaries.

Build studio
In small groups, participants refine the problem statement and value proposition, develop a stakeholder map, identify key risks and assumptions, define success metrics, and scope a realistic MVP that is narrow enough to build and meaningful enough to justify a pilot.

Outputs
Validated problem statement; stakeholder map; value proposition; risk and assumption log; success metrics; MVP scope.

Inter-session sprint

Between Week 1 and Week 2, participants refine their concept using course templates, faculty feedback, and lightweight stakeholder input from their own setting. They also complete any required build-environment checks so that Week 2 can focus on product specification and prototyping rather than technical setup.

Sprint submission: Refined problem statement, MVP scope, success metrics, and key assumptions.

Day 3: Turn the concept into a buildable product

Objective: Translate the validated concept into a buildable specification and develop a prototype v1.

Faculty session
Product Requirements, Architecture, and Agile Execution for Health Care AI (Panch and Mauro)
Participants learn how to move from concept to product definition, including what a Product Requirements Document (PRD) is, why architecture matters, how to specify workflows and acceptance criteria, and how human oversight and evaluation need to be designed into the product from the start.

Guided lab
From Concept to Product Specification (Technical faculty)
Participants develop the core components of a PRD: user personas, user stories, workflows, acceptance criteria, success metrics, data needs, architecture overview, and guardrails. The session also introduces agile development practices suitable for AI-enabled health care products.

Build studio
Working in small groups, participants build a prototype v1 using AI-assisted development tools, GitHub, and structured sprint methods. Faculty provide feedback on product scope, architecture, usability, and implementation readiness.

Outputs
PRD; workflow design; architecture overview; success metrics; guardrails and evaluation plan; prototype v1.

Day 4: Pilot, pitch, and next steps

Objective
Convert the prototype into a pilot-ready proposal with a clear pathway to adoption, funding, and scale.

Faculty session
From Prototype to Pilot and Scale (Atun)
Participants learn how innovations move from working prototype to organizational adoption. The session covers implementation pathways, governance, operating model considerations, evidence generation, and how solutions scale in health care organizations and public-sector settings.

Guided lab
Pathways to Adoption, Funding, and Scale (Technical faculty)
Participants refine their prototype, shape the launch pathway appropriate to their context, and structure a pitch tailored to one of three audiences: internal innovation leadership, venture or external funding, or public-sector decision-makers.

Build studio
Participants finalize the prototype, build the pilot or implementation roadmap, prepare a concise 90-day action plan, and rehearse the final presentation.

Final presentations
Each participant or team presents a live demo and a decision-ready pitch, followed by structured faculty feedback.

Outputs:Functional prototype or early MVP; pitch deck; pilot or implementation roadmap; governance and adoption considerations; 90-day action plan.

This program is designed for leaders with foundational AI fluency who are responsible for innovation, digital transformation, product strategy, clinical operations, venture creation, or public-sector modernization.

It is especially relevant for chief innovation, digital, information, medical, and strategy leaders; clinical and operational leaders sponsoring AI-enabled workflow redesign; product leaders and founders building health care AI products or services; and payer, life sciences, and public-sector leaders developing new AI-enabled offerings or initiatives.

Participants may attend individually or in small teams. No formal software engineering background is required, but participants should be comfortable experimenting with AI-assisted tools in a guided environment.

Trishan Panch, MD, MPH is the CEO of LUNRStudio, Executive Chair and Chief Strategy Officer at Lumin Health, Past-President at Harvard TH Chan School or Public Health Alumni Association and Board Chair at Healthcare for All. He co-founded Wellframe and co-led the company from inception in 2012 to acquisition by Blackstone/Health Edge in 2021 serving as Chief Medical Officer, Chief Product Officer and Chief Innovation Officer. He is the holder of the patent on the Wellframe technology and is the inaugural recipient of Harvard’s Public Health Innovation Award for his work in Digital Health.

 
Rifat Atun, MD, FRCP, FFPH, FMedSci is Professor of Global Health Systems and the Director of The Health System Innovation Lab at Harvard University. He is Vice Dean for Non-Degree Education and Innovation at Harvard T.H. Chan School of Public Health. In 2026, he was appointed to an endowed professorship as Julio Frenk Professor of Public Health Leadership.

All participants will receive a Certificate of Participation upon completion of the program. This program also contributes to the AI in Health Care Certificate of Specialization and the AI in Health Care Leadership Advanced Certificate of Specialization. Click here for more information.

Before the first live session, participants should identify one candidate problem or opportunity area from their organization or venture context. They should also complete a short orientation covering the Healthcare AI Studio, course expectations, and any required technical setup for the build sessions. Participants attending as teams should come prepared to work on a single project throughout the program.

The most important structural change is this: every day should have one dominant job. Once you make that explicit, the course stops feeling slightly hybrid and starts reading like a disciplined build sequence. This version is much closer to something you could put on the site.