Claude Code
Anthropic's agentic coding CLI — the spine of our AI-augmented research and education workflows. We use it, teach it, and build on it.
Research & Outputs
What we've shipped, and what's next
Open work items produced across the institute's programs. Live platforms lead; shipped open resources, the research stack we build on, and items in preparation follow.
Shipped outputs
4
Syllabi & outlines
2
Live platforms
Open in a browser today
Three deployed platforms you can click into right now — the products anchoring our active programs.
Using… · Platform
FIH Scout ↗
Oncology first-in-human trial intelligence — a live web platform exploring AI-assisted investigator and site ranking for early-phase oncology studies. Open for academic collaborators to try.
Open the platform →
Using… · Platform
Northern AI Hub ↗
Five-week in-person certificate program, AI in Production for Small Business, hosted on a partner campus in Middletown, NY — open application site with program structure, cohort timeline, and enrollment flow.
Open the platform →
Using… · Platform
AI in Health Care — From Models to Medicine ↗
Course platform co-developed with a clinical affiliate and our higher-education partner. A ten-week three-track curriculum covering clinical AI from models to bedside; open website with full syllabus, research papers, and partner notes.
Open the platform →
Open resources
Courses, keynotes, and tools under open license
- Keynote
My Daily Use of Claude Code — a doctor's AI-augmented workflow ↗
Full recording of the keynote, illustrating concretely how a practicing clinician integrates Claude Code and AI agents into daily clinical and research work. Hosted on Google Drive; open for anyone with the link.
From Pillar 3 · Using…
Research stack
The tools and databases we build on
A layered stack — frontier language models at the top, the open-source ecosystem beneath, and curated oncology and biomedical databases anchoring every claim.
Gemini 3.1 Pro
Google DeepMind's long-context multimodal model — used for medical-image review, large-document synthesis, and cross-checking against Claude outputs.
GPT 5.4
OpenAI's frontier model — used as a second opinion across reasoning tasks and for multi-model voting when a single answer is not enough.
GitHub open-source tools
The open-source ecosystem — Python / TypeScript libraries, statistical toolkits, medical-imaging SDKs — assembled into reproducible pipelines.
PubMed
NLM's biomedical literature database — the canonical source for systematic reviews, meta-analyses, and evidence mapping across oncology and urology.
Oncology society databases
ASCO, ESMO, AACR, SITC conference abstracts and clinical-practice guidelines — the working knowledge base for first-in-human trial context and biomarker strategy.
Syllabi & outlines
Public roadmaps — outline first, full content next
Course and manuscript outlines published before the full material, so collaborators and reviewers can shape the direction early.
- Syllabus outline
FIH Trial-Design Methods Outline
A public roadmap for the methodological manuscript on AI-assisted Phase I/II oncology trial design. We publish the outline before the paper so collaborators and reviewers can engage with the methodology early.
- Problem framing — where current Phase I/II trial design decisions lose efficiency
- AI-assisted investigator and site selection methodology
- Starting-dose cohort allocation under sparse prior information
- Biomarker strata identification and validation approach
- Open-benchmark protocol — data, code, and negative-result release policy
- Syllabus outline
Clinician × AI Co-taught Curriculum Outline
Draft syllabus of the co-taught module series developed with our higher-education partner — concept scoping, weekly topics, assessment approach, and reuse notes for other accredited institutions.
- Module 1 — Foundations: how clinicians already use AI (often without naming it)
- Module 2 — Critical appraisal of medical-AI literature and marketing claims
- Module 3 — Hands-on AI tooling in clinical workflow scenarios
- Module 4 — Ethics, bias, and patient-facing communication
- Assessment — case portfolios; reuse notes for partner institutions
In preparation
Active work items
To be released once board-approved.
- Research preprint
Medical-school feasibility study (in preparation)
Regulatory-landscape analysis (LCME, NYSED, NECHE) informing a partner institution's medical-school planning. Targeted for release after board-approved redaction.
From Pillar 2 · Using…