Projects

StataAgent

A specialized agent model that integrates with Stata for natural-language data analysis. Ask questions like "What is the average income of individuals in 2022?" and the agent queries Stata and returns a structured answer.

Explain — LLM Package for Stata

A Stata package that integrates directly with large language models to assist with code debugging, explanation, and improvements. Supports both hosted APIs and locally-run models, letting researchers get inline AI assistance without leaving their Stata workflow.

StataHelper — Parallelized Stata Workflows

A Python package that simplifies the PyStata interface and enables parallelized Stata execution. Especially useful for large-scale data tasks where vanilla Stata's single-threaded model becomes a bottleneck.

Large-scale Text Classification with ANNOY + SLM

Demonstrates how to combine lightweight SLM embeddings with the ANNOY (Approximate Nearest Neighbors) library for efficient multi-class text classification at scale — a practical alternative to fine-tuning large models for classification tasks.