The fractional, forward deployed CTO. One person, two jobs, done in the codebase.

Most healthcare operators we talk to need two people they can't hire. They need a senior technical voice in the room with the board, the clinical lead, and the payer. They also need someone who can ship the actual software, in the actual codebase, next week. The market has invented separate roles for each. We think that's the wrong unit of decomposition.

The fractional CTO market is real, and most of it is advisory. A calendar of one-hour meetings, a slide on a strategy doc, an opinion on a hire. Useful, but a step removed from the codebase where the actual risk lives. The advice is correct in the abstract and stale by the time it lands in a pull request.

The forward deployed engineering market, the Palantir lineage, is also real. A high-hour senior engineer dropped inside the operator's repo, on the operator's standup, writing production code. Excellent on execution. Weak on the room where the next quarter gets decided.

The fractional advisor knows what the company should do. The forward deployed engineer knows what the code is doing. The healthcare operator pays twice and still doesn't have a CTO.

The synthesis

The fractional, forward deployed CTO is one senior who does both jobs. Same person writes the merge commit and sits across from the board. Same person reads the SOC 2 finding and pairs on the fix. Same person picks the model and the model evals and the release gate.

This is not a hybrid because it splits time differently. It's a hybrid because the two halves need each other to be correct. A roadmap written without code context drifts into consultant-speak. Code written without roadmap context optimizes for the wrong horizon. The hand-off between strategy and shipping is the single most expensive seam in healthtech operations. Removing the seam is the whole pitch.

What an engagement actually looks like

A normal week, in shape:

1. Monday. Standup with the engineering team. Triage the queue. Pair on whatever is blocking the senior engineer who's on the workflow this week.
2. Tuesday. Architecture read on the next quarter. Memo to founders. Vendor call with a payer engineering team.
3. Wednesday. Production code. PRs against the workflow that matters this month. Reviewer of record on everything that touches PHI or billing.
4. Thursday. Board call or clinical leadership sync. Eval review with the safety team. Hiring loop for the first or second in-house engineer.
5. Friday. Eval run on the agentic workflow. Postmortem on whatever incident happened. Written summary to the operator. No theater, no slideware.

Half the time is in the codebase. Half the time is in the room. The artifacts produced are runnable code, written memos, and measured outcomes against a metric the operator already runs against.

Why healthcare needs this specifically

Three forcing functions, none of which exist as cleanly in other verticals.

• The compliance envelope is a code property. HIPAA, SOC 2 Type II, and BAA scoping are not a slide deck. They show up as audit logs, encryption keys, row-level access, and PHI boundaries in the schema. A CTO who only advises cannot defend an audit. A CTO who only codes cannot frame the audit for legal and the board.
• Clinical workflows are unforgiving. Software that ships into a clinic gets used by clinicians on shift, under load, with patients in the room. Architectural decisions made one step removed from that constraint break in predictable ways. The CTO needs to feel the failure modes.
• EHR interop is brownfield. Epic, Oracle Cerner, Athenahealth, eClinicalWorks, AdvancedMD, and the long tail of practice-shaped systems are real, persistent, and quirky. The work is not a clean platform build, it is a decade-long integration with named systems and named quirks. An advisor who has never written against FHIR R4 or HL7v2 against a real interface engine has no view on how long it actually takes.

When this is the right call

The fractional, forward deployed CTO fits a specific window in a healthcare company's life.

1. Post-seed, pre-Series A. The product exists, real users are on it, and the founders need senior technical authority without committing to a full-time hire at the wrong salary band.
2. Brownfield modernization. A twenty-year-old monolith, a working business, a board that wants AI-native software without a replatform. The CTO needs to be inside the monolith and inside the board meeting in the same week.
3. Pre-clinical launch. The team has a working prototype, a clinical pilot is on the calendar, and there is no senior person who can own both the safety architecture and the conversation with the medical director.
4. Bridge between full-time CTOs. A CTO has left or is leaving. The company needs to keep shipping, keep recruiting, and not lose six months to a search.

When it isn't

We don't take the engagement when a full-time CTO is the obvious right answer and the company can hire one. We don't take the engagement when the team needs a long-tenured employee for cultural reasons that can't be served by an embedded partner. We don't take the engagement when the operator is looking for an advisor disconnected from the code. That market exists. It isn't us.

The shape of the deal

Quarterly retainer. Senior throughout. The person on the kickoff call is the person writing the merge commit. You retain everything we ship, source, infra, evals, hand-off docs. No perpetual license, no platform tax. When the engagement closes, it closes cleanly.

A fractional, forward deployed CTO is not half of an advisor and half of an engineer. It is one senior whose job description includes both, because in healthcare neither one of those works alone.

If you want to test the model, the cheapest way is a two-week architecture read. One senior, one written deliverable, no team commitment yet. That's usually enough to know whether the embedded build is the right next step, or whether the answer is something else entirely.