Senior Statistical Programmer · CDISC Automation

Automation for regulated clinical data.

I turn manual clinical-programming checks into tested, auditable Python tools. SDTM, ADaM and TFL work, built for locked-down GxP environments.

View the work Download résumé
Experience10+ years GxP
FocusSDTM · ADaM · TFL
StackPython · SAS · CDISC
Tools built8 production tools
// the pipeline my tools cover
RAW / EDC→ SDTM→ ADaM→ TFL→ DEFINE

Each tool automates a step on the way from collected trial data to a regulatory submission.

Projects

8 tools · Python · Streamlit
CIS Clinical Integrity Suite (CIS) A programmer-facing SDTM/ADaM data quality checker that runs locally before P21. PythonStreamlitpandas v34.2 View details → MMS SDTM Master Mapping System Raw-variable-first, master-guided automated mapping of raw study data to CDISC SDTM, with a SAS generator. Pythonscikit-learnStreamlit v28.13 View details → SCV SDTM Completeness Layered traceability and completeness validation from raw through SDTM, with a live dashboard. PythonStreamlitpandas v12 (0.4.0) View details → SVT Smart Validation Tool Pre-delivery QC for SDTM/ADaM/TFL: scans logs and listings, scores health, and flags gaps. PythonTyperStreamlit v0.13.1 View details → CDT Clinical Data Toolkit (CDT) A multi-tool Streamlit workbench: figure QC, TFL version compare, and dataset utilities. PythonStreamlitpandas v25 View details → EXP SDTM Clinical Data Explorer An interactive SDTM dashboard: per-subject drill-down, CSR-quality TFLs, and a knowledge graph. PythonStreamlitPlotly v5 View details → CTI Clinical Trial Intelligence Platform (CTI) Semantic extraction of TFL programming specs from SAP documents — pattern-based, no LLM. PythonStreamlitpython-docx v6 View details → GRA ADaM Knowledge Graph A knowledge graph linking every ADaM variable back to its SDTM source, derivation, and SAP evidence — no LLM. PythonStreamlitnetworkx v3.1 View details →

// where this is heading

From a set of tools toward one platform

All eight tools share the same low-level work: reading SAS, XPT and Excel, and writing Excel and HTML reports. The direction is to factor that shared foundation into a single internal library — cdisc-core — so every tool draws its parsers, validators and reporters from one tested, versioned base instead of its own copy. That is what turns a collection of tools into a platform.

About

Study Lead and Senior Statistical Programmer building automation that makes locked-down GxP environments faster without breaking traceability. Deep CDISC work in SAS and Python across oncology, virology, cardiovascular and early-phase trials.

Read more →
10+
Years GxP
8
Tools built
5
Therapy areas
CDISC
SDTM · ADaM · TFL