James H Wade

Research scientist building AI developer tools for industrial R&D. Writing about R, Shiny, and large language models.

About Me

Latest Posts

Python in Your Browser with Pyodide
Python
WebAssembly

Pyodide runs CPython in the browser via WebAssembly. Edit and run every example on this page: no server, no install.

James H Wade
Feb 21, 2026

Turning Shiny Apps into MCP Apps with shinymcp
Shiny
MCP
AI
R
ellmer

shinymcp converts Shiny apps into MCP Apps: interactive UIs that render directly inside AI chat interfaces like Claude Desktop.

James H Wade
Feb 21, 2026

Disposable Shiny Apps
Shiny
AI
posit::conf
R

Build more Shiny apps. Then throw them away.

James H Wade
Nov 3, 2025

Disposable Shiny Apps: Annotated Talk Notes
Shiny
AI
posit::conf
R

My posit::conf(2025) talk, slide by slide — what I was trying to say, what worked, and what I’d change.

James H Wade
Sep 18, 2025

R in Your Browser with WebR
R
WebAssembly

WebR brings R to the browser via WebAssembly. Every code cell on this page runs locally: edit it, run it, break it.

James H Wade
Aug 13, 2023

Teaching ChatGPT What It Doesn’t Know
ChatGPT
LLM
NLP
Web Scraping
R
OpenAI

Using a retriever (or vector database) to provide missing context to ChatGPT, similar to ChatGPT Retrieval plugin

James H Wade
Mar 25, 2023

Exploring {dm} Alone
database
RDBMS
data management
R
dm
TidyTuesday

Using data from Alone to test drive the {dm} package as part of week 4 of TidyTuesday

James H Wade
Jan 25, 2023

During the MLOps cycle, we collect data, understand and clean the data, train and evaluate a model, deploy the model, and monitor the deployed model. Monitoring can then lead back to collecting more data. There are many great tools available to understand clean data (like pandas and the tidyverse) and to build models (like tidymodels and scikit-learn). Use the vetiver framework to deploy and monitor your models.

MLOps: Moving from Posit Connect to Azure
mlops
vetiver
pins
deployment
R
cloud
Azure

Combining Posit’s open source tools and Azure for a cloudy MLOps deployment

James H Wade
Jan 22, 2023

Bayesian Optimizaiton with Tidymodels
machine learning
modeling
tune
deep learning
torch
R

Model tuning or torch models with Bayesian optimization using tune, workflows, brulee, and other friends from tidymodels

James H Wade
Jan 1, 2023

MLOps: The Whole Game
mlops
modeling
vetiver
pins
deployment
R

An example of model building, model deployment, and model monitoring with R using palmerpenguins

James H Wade
Dec 27, 2022

How to Teach Tech
learning
Much of my inspiration for this project come from Greg Wilson, founder of the Carpentries. These notes are based on his talks about how to teach tech. The first is a talk he…
James H Wade
Feb 1, 2022

Notes on learning
learning
I have an idea for lowering the barrier to learn new R packages. I have built a few private packages, but nothing of mine fits the bill for what I want to try. There are…
James H Wade
Jan 30, 2022
No matching items