The Medicus Tiro Monthly

March was the most active month we've had. Demos every week. The GME Manager Monthly subscriber list doubled. And the market signal is unmistakable: Family Medicine programs know the tools they have aren't built for what they're being asked to do, and they're actively looking for what comes next.

That's the environment we're building into. Here's where things stand.

Last month we noted that we expected to announce our initial investor in this edition.

We are still on track. The conversation is progressing and the terms are aligned. We will share the announcement as soon as the paperwork is finalized — we expect that to happen before the next edition.

We are grateful for the patience and confidence of everyone tracking this closely.

Physician burnout is a well-documented crisis. What’s less understood is that every measure is worse — often dramatically — for the physicians still in training. The decade from the first year of medical school through the first year of independent practice isn’t just hard. It is the least supported stretch of a physician’s career. And this is not a selection problem.

Medical students enter training with better mental health and lower burnout rates than their age-matched college-graduate peers. By the second year of training, that advantage flips — and never recovers at any point during the remaining years of the decade. The training itself inflicts the damage.

•       Burnout: 47–50% of residents and fellows, vs. 43% of practicing physicians. Spikes to 45% during PGY-2 and PGY-3, when clinical responsibility scales up and structured supervision scales down.

•       Depression: 27% of medical students screen positive — more than double the rate of age-matched peers. Medical students are 3x more likely to die by suicide than age-matched controls.

•       Work-life satisfaction: 34% of residents, vs. 42% of attendings — and vs. 58% of the general U.S. workforce.

•       Debt-to-income: Average medical education debt exceeds $223,000, serviced on resident salaries of ~$67,000 — the most extreme ratio of any profession at any career stage, sustained for 3–7 years.

•       The help-seeking “penalty” culture: Only 16% of medical students experiencing depression pursue treatment — because doing so can jeopardize licensure and career advancement.

•       Attrition: 30–40% of new physicians leave their first job within five years, citing a lack of mentorship and support in the transition to independent practice.

This isn’t just a physician problem. Resident burnout is directly associated with higher medical-error rates, increased attrition, and career regret — 14% of PGY-2 residents overall, and up to 33% in some specialties, report they would not choose medicine again. The damage extends to every patient they see, every system they work in, and every dollar invested in producing them.

These numbers improve after training. Burnout drops. Satisfaction rises. Mid- and late-career physicians consistently report better outcomes — confirming that the crisis is concentrated in this decade, not distributed across a career.

That’s the reality. Here’s what we’re building to change it.

We haven’t told the product story directly in this newsletter yet, so April is the right moment.

Medicus Tiro is building purpose-built AI agents for every stage from the first year of medical school through the first year of independent practice. Each agent solves a specific problem. All of them read from and write to a shared data layer — the Lifelong Portfolio — that eliminates the silos that have fragmented every stage of the physician journey until now. For programs, it’s the connective tissue that makes every agent smarter. For residents, it’s something more personal: a verified, portable record of the competency they spent a decade building — a foundation for a career, just as their education is a foundation for their practice.

This is why the architecture matters as much as any individual feature. When a new use case emerges, we don’t rebuild the intelligence — we give it a new surface and build a purpose-built application around it. That is the capital efficiency story. That is why the pace is possible.

Today, the first six products are live or in development. They are the starting point, not the finish line:

GME Manager — Core competency-based residency management: milestone tracking, CCC meetings, AI-drafted evaluations, ACGME data exports, duty-hour compliance, procedure logging. The flagship.

GME Manager Mobile — A fully native mobile interface for the same platform. 98.3% of residents use smartphones in clinical practice. We built for the phone first — Voice Logging, Shake to Log, biometric security, full offline support.

GME Onboarding Manager — Automates the Match Day to Day One window. Self-configures from specialty, state, and class list. Runs itself March through July.

GME AI Advisor — Agentic AI and predictive analytics that coexists alongside legacy platforms — New Innovations, MedHub — without disrupting a single resident workflow.

GME Recruiting Manager — Transforms raw ERAS data into AI-evaluated applicant profiles, with Variable Autonomy™ controls at every stage of screening, interviewing, and ranking.

GME Schedule Manager — The first shift scheduler built for GME. Every assignment enforces 8 ACGME duty hour rules in real time — then optimizes for training fit, continuity, and resident wellness.

GME Manager is the flagship and first to market. We’re launching in Family Medicine — 800+ accredited programs where competency-based evaluation demands are acute and existing tools are weakest. But the vision doesn’t stop at residency, and it doesn’t stop at six products. The same shared data layer powers every agent we build — including surfaces we haven’t announced yet. FM is where we prove the model. The Portfolio is how it all compounds — one compounding data moat that grows with every program we serve. And soon, a new way in — GME Manager Agent Studio.

The architecture above — purpose-built agents, one shared data layer — describes how we build. What follows is a preview of how programs will buy.

We are developing GME Manager Agent Studio — an additional way for programs to access GME Manager. The full platform remains our primary offering. Agent Studio gives programs a second entry path: the ability to start with a single purpose-built AI agent and grow into the full experience over time. Each agent solves a specific problem on its own. Together, they compound.

The idea is straightforward: some programs are ready for the full GME Manager platform today. Others want to start with the problem that’s most urgent. Agent Studio gives them that option. A Program Director facing the ABFM June deadline could activate the Attestation Agent — pre-loaded with all 15 ABFM Core Competencies — and be running it against their residents within an hour. A credit card decision, not a procurement decision.

And then the architecture does its work. Every agent reads from and writes to a shared data layer. The Attestation Agent creates evidence records. When the same program later activates the CCC Agent, that data is already there — pre-populating milestone suggestions, enriching CCC prep cards. Each agent makes every other agent smarter. That’s not a bundle. That’s a flywheel.

The platform foundation — Action Rail™, Insight Rail™, Variable Autonomy™, the 12 pre-loaded clinical data sets, and the Lifelong Portfolio — ships with every agent. Programs that start with one agent and grow into the full suite get the complete GME Manager experience. The destination is the same. Agent Studio is simply another way to get there.

For investors: Agent Studio adds a second go-to-market path alongside the full platform sale. Programs ready for comprehensive residency management adopt GME Manager directly. Programs that need to start smaller enter through a single agent and expand as each agent compounds the value of the others. Two entry paths, one destination, one compounding data moat.

More details next month.

One detail worth drawing attention to in the prototype: on the GME Manager login screen, alongside institutional SSO, residents and faculty can sign in with their hStream ID — the verified healthcare identity credential managed by HealthStream, used across thousands of health systems for credentialing, learning management, and workforce verification.

This is not incidental. hStream ID is the verified identity layer used across thousands of health systems — and that integration is a distribution advantage. Rather than building a competing identity layer and asking institutions to adopt it, we meet physicians and programs inside the infrastructure they already use. That changes the adoption conversation entirely.

For investors: hStream integration means GME Manager is already inside the institutional infrastructure that governs thousands of health systems — not requesting access to it.

We have made a deliberate architectural commitment that has implications well beyond compliance: every Medicus Tiro product is designed as a No-PHI system. We never store, process, or transmit Protected Health Information. We do not require a Business Associate Agreement.

This is a strategic choice, not a limitation.

Healthcare software that touches anything health-adjacent typically requires institutions to execute a BAA before deployment — triggering legal and compliance review that can add weeks or months to a sales cycle, and frequently stops pilots in their tracks at smaller programs without dedicated legal teams. We designed our way out of that friction entirely.

Here's the architecture in brief. Patient encounters imported for Continuity of Care tracking are de-identified at the ingestion boundary — the system receives age/sex tokens, visit category, E&M codes, and dates, never patient names, MRNs, diagnoses, or ICD codes. Resident health information is compliance-status-only: three fields per requirement, no clinical detail. The Wellness module is architecturally anonymous — no user ID, session token, or granular timestamp is stored with any response, making re-identification technically impossible at the database level. CCC and remediation records are governed by FERPA as educational and employment records, not HIPAA, with persistent in-application guardrails to keep them that way.

For investors evaluating the sales motion, the implications are direct:

No BAA means shorter cycles. Programs can evaluate, pilot, and deploy without triggering institutional legal review. That is particularly meaningful for the 800+ Family Medicine programs we are targeting first — many of which are community-based, without large compliance teams — and equally meaningful for every market we enter next.

No BAA means broader distribution. Products that require BAAs face a gatekeeping layer that No-PHI products skip. LaunchPad pilots — our self-guided 90-day model — work precisely because there is no legal prerequisite to starting.

No BAA means a cleaner data story. The intelligence our platform generates is about physician development, not patient health. That distinction matters for how the data compounding story is told, how the platform is positioned in enterprise conversations, and how the data moat is ultimately valued.

We believe this architectural decision will look increasingly prescient as healthcare AI scales and institutions become more cautious about PHI exposure in AI-adjacent systems.

One of the questions we hear from investors is how a platform selling into fragmented markets — starting with 800+ Family Medicine programs and expanding across the full physician journey — converts interest into revenue without an expensive sales and implementation machine. The answer is in the model itself.

Fast entry. Every product and every agent ships with a free 90-day LaunchPad Pilot — self-guided, AI-configured, zero implementation staff required. Programs enter their specialty, state, and class list. The platform configures itself. There is no 12-week onboarding, no project manager, no professional services engagement. We believe if your product needs an implementation team, your product isn't good enough. The No-PHI / No-BAA architecture removes the last remaining friction point — programs can start without triggering legal review.

Fast revenue recognition. The pilot is free. The conversion is not. Programs that complete the 90-day LaunchPad move to a hybrid subscription: a base platform fee plus a usage component that scales with the AI capabilities they adopt. Revenue recognition begins at conversion, not at contract signature. Because pilots are self-serve and parallel — we can run dozens simultaneously — conversion pipeline builds quickly without proportional headcount.

Embedded scalability. The usage component is the long game. In GME Manager, programs that start with core milestone tracking and AI-drafted evaluations grow into CoC Predictor, APE Report generation, CCC Meeting Mode, ABFM Attestation — each incrementally adopted, each incrementally billed. The same expansion pattern applies to every product we build. Agent Studio adds a second axis: programs that enter through a single agent expand as the shared data layer makes each additional agent more valuable. A program that entered at the base tier in Year 1 is a materially larger revenue relationship by Year 3, without a new sales motion. The architecture that powers a growing family of products from a shared data layer means new surfaces — Recruiting Manager, Schedule Manager, and future products beyond residency — extend the same revenue relationship rather than requiring a new one.

The model aligns our incentives with program outcomes: we grow when they use more, which happens when the platform is delivering value. That is the business we are building.

The most important decade of a physician’s career. The least supported. That’s the problem we built this company to solve.

Agentic AI that follows physicians from the first year of medical school through the first year of independent practice. Purpose-built agents. One shared data layer. A portfolio the physician owns for life. A self-service Agent Studio that lets any program start with one problem today and grow into the full platform over time. Data that compounds. Trust that’s earned. A relationship that lasts 35 years.

If that resonates — whether you’re an investor, an enterprise platform, or a leader in medical education — join us. We are on a mission to make the least supported decade the most supported decade.

[email protected] | 720-352-5203
medicustiro.ai

And if you haven't subscribed to GME Manager Monthly — our product-focused newsletter for Program Directors and Coordinators — it's the fastest way to see what we're building at ground level. The subscriber list doubled in March.  Subscribe to GME Manager Monthly →

Michael Sousa
Founder & CEO, Medicus Tiro Inc.

Know someone who'd enjoy this? Forward it to a friend.