3. Data, Science & AI

Creates OpenAI-powered AI agents with tool-calling, multi-step reasoning, and autonomous workflows for complex task automation.

Builds RAG systems for document Q&A, knowledge bases, and semantic search using vector databases and LLMs.

Queries the CHGIS Temporal Gazetteer API for historical Chinese placenames, administrative units, and geographic data from 222 BCE to 1911 CE.

Designs and implements scalable data pipelines, ETL/ELT processes, and data infrastructure for efficient data processing and analytics.

Analyzes PostgreSQL usage in Rails applications to detect N+1 queries, missing indexes, and configuration issues, providing actionable optimization recommendations.

Manages end-to-end AI skill evolution via performance monitoring, variant generation, and data-driven optimization for continuous improvement.

Generates and interprets forest plots for meta-analysis using R's metafor package, visualizing pooled effects and confidence intervals.

Generates executable R code for meta-analysis workflows using metafor, including data preparation, model fitting, and visualization.

Comprehensive framework for validating data schema compliance, quality, and integrity across databases, APIs, and pipelines with anomaly detection and scorecards.

Provides on-device text recognition from images using iOS Vision Framework, supporting multiple languages and efficient image processing.

Teaches network meta-analysis (NMA) for comparing multiple treatments, including indirect comparisons, network plots, and league tables in evidence synthesis.

Analyzes file content, structure, and statistical metrics across multiple file formats.

Executes SQL queries on Snowflake data warehouse with multiple authentication methods, supporting JSON, table, and CSV outputs.

Teaches meta-analysis techniques for diagnostic test accuracy studies, covering sensitivity, specificity, SROC curves, and bivariate models with QUADAS-2 quality assessment.

AI-powered audio transcription and translation toolkit supporting 99 languages, subtitle generation, and batch processing for speech-to-text applications.

Converts markdown documentation and tutorials into interactive Jupyter notebooks by splitting content into cells and extracting code blocks.

Teaches Individual Patient Data (IPD) meta-analysis methods for combining raw participant-level data across studies to explore treatment effects and time-to-event outcomes.

Enables creation, editing, and analysis of spreadsheets with formula support, formatting, and data visualization for structured data processing.

Framework for implementing safety safeguards in autonomous systems, including kill switches, resource limits, and anomaly detection to prevent operational harm.

Teaches Bayesian meta-analysis techniques including prior specification, MCMC, and posterior interpretation for probabilistic data analysis with sparse datasets.

Detects publication bias in meta-analyses using funnel plots and Egger's test to assess missing study impacts on conclusions.

Assesses between-study heterogeneity in meta-analyses using I², Q statistic, tau², and prediction intervals to evaluate consistency and pooling appropriateness.

Configures data sources, ensures reproducibility, and enables regeneration for AI/ML data pipelines.

Extracts and prepares study data for meta-analysis, calculating effect sizes, estimating variances, and handling missing data to convert statistics into analyzable formats.