Postdoctoral Scholar, Clinical AI & Agentic Systems (Serghiou Lab)

Faculty Sponsor First name: Stelios Faculty Sponsor Last Name: Serghiou Stanford Departments and Centers: Epidemiology and Population Health Postdoc Appointment Term: Full-time, yearly renewal. Start: January (preferred) to September 2027.

Appointment Start Date: January 2027 (preferred) Group or Departmental Website: https://serghioulab.stanford.edu/ https://med.stanford.edu/content/sm/epidemiology-dept.html/ How to Submit Application Materials: Email all documents to Prof. Serghiou at serghioulab@stanford.edu . We review applications as they come. Does this position pay above the required minimum?: No.

The expected base pay for this position is the Stanford University required minimum for all postdoctoral scholars appointed through the Office of Postdoctoral Affairs. The FY27 minimum is $79, 056. Why this matters More than 900K US adults die every year from a heart attack or stroke.

Even though these are largely preventable, more than 60M Americans are not receiving simple life-saving medication for it. We are building the AI-driven prevention systems necessary to make sure no patient falls through the cracks again.

Who we are We are an interdisciplinary group of clinicians, engineers and epidemiologists, who are passionate about building and rigorously evaluating the clinical intelligence necessary to provide high quality care at unprecedented scale. We believe that care should be proactive, personalized, value-driven, and continuous, and are developing the technology necessary to make it possible.

What you'll build We're building agentic systems that turn patients' longitudinal records into proactive, evidence-based care. You'll lead a distinct line of inquiry within this program, from method development to real-world clinical evaluation. Your work will span: ● Medical records . Curate and make sense of longitudinal records for millions of patients. ● Language models .

Train and evaluate models capable of generating clinically relevant outputs. ● Conversational agentic systems . Build agents that can reason over records and collaborate. ● Dynamic dashboards . Show clinicians what matters, and what's missing, at a glance. ● Clinical collaborations . Define what matters with clinicians and test systems on real decisions. ● Communication .

Share insights through first-author papers, talks and open code. Your first year 1. First month. Settle in, meet the immediate and extended teams, and get access to data. 2. By three months . Basic infrastructure, early prototypes and proofs of concept. 3. By six months . First end-to-end prototype, an evaluation harness and a submitted abstract. 4. By twelve months .

Clinical real-world evaluations and a submitted first-author paper. What you get ● Real-world impact . Use real-world data to improve care in a real health system. ● Work that matters . Lead first-author research at the cutting edge of clinical AI. ● A hand in building the lab . As one of the first hires, shape how we work and who joins next. ● A dedicated mento r.

A mentor invested in your personal and professional development. ● A rigorous community . Join people who prize rigor, reproducibility and openness. Who might not enjoy this We're new, so if that sounds like fun rather than a warning, read on, as things will be scrappy before they are smooth. We also work hard: the bar is high and we take pride in creating exceptional work at record speed.

Finally, we work methodically and transparently to make sure our work is rigorous and reproducible.

Required

Qualifications

Who you are The bar : you've previously trained generative AI models, built safe and robust agentic systems over large data, created rigorous evaluations, and published your work. We also look for evidence of: ● Prior experience .

A PhD (completed or nearing completion) in computer science, engineering, data science, statistics, epidemiology or equivalent, and a first-author publication. ● Machine learning and AI . You've trained and applied modern ML and language models to real problems, from retrieval and extraction to open-ended outputs. ● Agentic engineering .

You've built agents that plan, call tools and check their own work, using robust, tested Python that have been deployed or contributed to a publication. ● Data engineering . You've built reliable pipelines over large, messy data with SQL, cloud compute, and versioned, reproducible workflows. ● Data science .

You've designed rigorous evaluations, quantified uncertainty, and made sense of your results using foundational principles of statistics and study design. ● Rigor and reproducibility . You have a track record of transparent work, such as shared code, shared data, pre-registration, or adherence to reporting guidelines. ● Communication .

You can write and speak clearly, communicate effectively with non-technical individuals, and demonstrate your ability to collaborate effectively with clinicians. Nice to have, not required : experience working with large longitudinal health data in secure environments (OMOP, FHIR, Epic), clinical NLP, imputation, entrepreneurship, or have an MD. Just as important is how you work .

We look for mission-driven individuals, who prioritize kindness, intellectual humility, collaboration, urgency, and dedication to going above and beyond. Required Application Materials: CV Your interest for the lab Your proudest first-author publication Note that a few genuine and non-AI generated sentences beat a polished cover letter. Stanford is an

equal opportunity employer and all qualified applicants will receive consideration without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, or any other characteristic protected by law. Menu Toggle menu visibility

How To Apply Open Postdoctoral Positions Finding a Faculty Mentor Cost of Living Housing Fellowships at Stanford Fellowships outside Stanford About

About Us Postdoc Benefits Postdoc Emergency Resources Verify Appointments Guidelines Policy Funding Guidelines Budgeting for Fellowships Contact Us Stanford School of Medicine E.D. Stone Edwards Building 300 Pasteur Drive, Suite 133 Palo Alto, CA 94304 Map to Building | Contact OPA Sunet Login