For medical students & residents

Publishable research for the match, in 6–10 weeks.

Probe Scholar takes you from a research question to a submission-ready manuscript you can defend in any interview. Real computational research, a named attending mentor, and work that is unmistakably yours.

Request beta access → See the pathway Now onboarding the first beta cohort
One question. Nine stations. A finished manuscript. Guided workflow · human-led
01
QuestionWeek 1
02
LiteratureWk 1–2
03
Study setupWk 2
04
Stress testWk 3
05
Mentor gateWk 3–4Approval
06
SimulationWk 4
07
ResultsWk 5–6
08
ManuscriptWk 6–9100% yours
09
SubmissionWk 9–10
Why it exists

The match now runs on publications most trainees can't reach.

Competitive specialties increasingly expect multiple publications, and the traditional path to them is blocked for the trainees who need them most.

~2×in four years

Publication counts expected for competitive specialties like dermatology, orthopedics, and neurosurgery have nearly doubled.

~30%of students

Only around a third of medical students have a research mentor. The rest depend on serendipitous connections.

12–18months

Traditional trainee research is gated by IRB, chart access, and mentor bandwidth long before a first result appears.

3–5×publication tax

International graduates and trainees at under-resourced institutions must publish far more to be competitive.

What it is

Legitimate computational research, compressed to weeks by removing everything except the thinking.

Probe Scholar
A guided research platform for medical students and residents that takes a trainee from a research question to a submission-ready manuscript in 6–10 weeks, running on Infiuss Health's Probe digital-twin simulation engine.
How the time is saved
Because studies run on simulated virtual patient cohorts, the recruitment, IRB, and data-collection stages that normally take 12–18 months disappear. What remains is the work that actually belongs to the trainee: forming the question, reading the literature, interpreting the findings, and writing.

AI is scaffolding throughout the workflow, never a replacement for the trainee's reasoning.

The product does not generate research for you

The pathway

Nine guided stations, two of them non-negotiable.

Every study moves through the same sequence. Two stations exist specifically to make the work defensible: the mentor approval gate, and the discussion section you write entirely yourself.

Station 01

Choose or generate a question

Browse a library of vetted research questions filtered to your specialty, or generate a custom one. Every AI-suggested hypothesis arrives with real, PubMed-validated citations you can open.

Station 02

Build a literature review you can defend

The workspace surfaces 20–40 relevant papers with summaries. You read them, take notes, and write your own synthesis of the field.

Station 03

Configure the study

Set cohort size, endpoints, and subgroups within the question's bounds. An inline advisor recommends the right statistical approach and explains it.

Station 04

Survive the stress test

AI raises the strongest counter-arguments a reviewer might make. You answer each one in writing. Those responses become part of your discussion and your armor in the interview.

Station 05 · Gate

Your mentor signs off

A named attending reviews your design and your stress-test answers, then approves it. Nothing runs until they do. This is the defensibility gate.

Station 06

Run the simulation

Your approved study runs on the Probe digital-twin engine against a virtual cohort. No real patient data is ever involved.

Station 07

Interpret every finding

Results arrive with charts and statistics, and a prompt for each finding: what does this mean clinically, what are its limits, what would you tell a program director. You write the answers.

Station 08 · Yours

Write the manuscript

A section editor auto-populates the mechanical parts, but the discussion section receives no AI help by policy. Every export shows an AI-assistance heat map.

Station 09

Submit and track

Log your journal or conference submissions and track them through review to a decision, with deadlines and reminders built in.

Human judgment stays human.
Question formulation, interpretation, writing, and peer-review response are never automated.

Defensibility

Built to answer the one question that matters: is it really yours?

The fear with any AI-assisted research is that it reads as slop the moment someone looks closely. Probe Scholar is engineered so that when someone does look closely, the work holds.

Senior author

A named attending vouches

Your mentor gates the design, reviews the manuscript, and signs on as senior author of record. Real accountability, not a rubber stamp.

Answered in advance

The stress test

You've already written responses to the hardest reviewer objections, reviewed by your mentor, before you ever walk into an interview.

Full transparency

AI-assistance heat map

Every manuscript export discloses exactly which text came from you and which came from AI, section by section.

Immutable record

Complete audit trail

Every action you and the AI take is timestamped and logged, exportable per study, so the provenance of the work is never in doubt.

No hallucinations

Citation grounding

Every AI citation is validated against PubMed or CrossRef and dropped if it can't be verified, before it ever reaches you.

No patient data

Virtual cohorts only

Studies run on simulated digital twins, so there's no PHI to protect and no privacy exposure to explain.

"What part of this is genuinely your work?"

The discussion section. Every word. By policy, it receives no AI assistance at all.

Who it's for

Two sides of the same project.

For trainees

You own the project, end to end.

  • Medical students (MS2–MS4) and residents applying to competitive specialties
  • Trainees without a research mentor, matched into a curated network instead
  • International graduates and students at under-resourced institutions
  • Anyone who needs defensible publications and doesn't have 18 months to wait
For mentors

High-leverage supervision, real authorship.

  • Curated network of attending physicians, onboarded and verified by Infiuss
  • Approve a well-designed study from your phone in about two minutes
  • Review manuscripts with the AI-assistance heat map in full view
  • Senior author of record on every project you supervise
What you walk away with

Not just a line on a CV.

A finished manuscript

Submission-ready, with auto-populated methods and a disclosure block, exportable as .docx.

A defensible lit review

Papers you actually read and synthesized in your own words, ready for the interview question.

Interview-ready answers

Written responses to the hardest objections about your work, mentor-reviewed in advance.

A named senior author

An attending who reviewed the work, gated it, and put their name behind it.

Questions answered

Probe Scholar, in plain terms.

What is Probe Scholar?

Probe Scholar is a guided research platform for medical students and residents who need publishable research for residency and fellowship applications. It takes you from a research question to a submission-ready manuscript in 6–10 weeks, running on Infiuss Health's Probe digital-twin simulation engine and supervised by a named attending mentor.

Does Probe Scholar just write my research with AI?

No. Probe Scholar does not generate research automatically. AI is used as scaffolding for literature synthesis, hypothesis critique, and drafting, but you formulate the question, interpret the results, and write the manuscript. The discussion section receives no AI assistance by policy, and every export includes an AI-assistance heat map showing exactly what came from you versus AI.

How long does a project take?

You can go from selecting a research question to a submission-ready abstract in 6–10 weeks of real work. Traditional trainee research is typically gated for 12–18 months by recruitment, IRB, and data collection; Probe Scholar removes those bottlenecks because studies run on simulated virtual cohorts.

Do I need my own research mentor?

No. Probe Scholar provides a curated network of attending mentors. A mentor approves your study design before any simulation runs, reviews your manuscript, and serves as senior author of record. It is built for the roughly 70% of medical students who do not have a research mentor.

Does it use real patient data?

No. Simulations run on virtual patient cohorts using the Probe digital-twin engine, so no protected health information is involved. No PHI is sent to any AI provider, and providers are configured for zero-retention or short-retention where supported.

Will program directors see this work as legitimate?

Probe Scholar is built for defensibility. A named attending mentor gates and co-authors the work, a hypothesis stress test forces you to answer the hardest reviewer questions in writing, an immutable audit trail logs every action, and every AI citation is validated against PubMed or CrossRef before you see it. The discussion section is written entirely by you.

What does it cost?

Probe Scholar is currently onboarding its first beta cohort with free access; broader pricing will follow. Request beta access to be considered for the initial group.

Beta cohort now open

Start your first defensible publication.

We're onboarding the first group of medical trainees and attending mentors. If you need publishable research for the match, or you want to mentor the students who do, get in touch.

Request beta access →