Tags

Browse writing by topic, from AI agent systems and local-first tools to browser-based AI and computer vision.
Local-First

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Project

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

LAPOP Data Dashboard

A dashboard prototype built with Shiny R

Cashbox Magazine Song Chart Analysis

A simple example of a data analysis pipeline

Think Strategically and Get Jobs - H1B Visa Analysis

Helping international data science students strategize their job search

What makes a trending video?

An analysis of trending YouTube videos

Computer Vision

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Web Application

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

LAPOP Data Dashboard

A dashboard prototype built with Shiny R

JavaScript

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Software Architecture

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Productivity Tool

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

Image Processing

Slides Thief 2.0: From Rectangle Heuristics to Evidence-Aware Review

A browser-local redesign of boundary detection, confidence, and batch geometry.

AI

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

Darknet to Keras - Model Transformation

New version based on TensorFlow 2.0

Agent

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Agent Memory

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Developer Tools

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

Data Governance

Designing Memory That Can Safely Forget

Persistent project memory is trustworthy only when forgetting is explicit, previewable, and bounded.

Git

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

Markdown

Why Project Memory Should Be Plain Text and Repo-Native

Durable coding-agent memory should live with the code, survive the tool, and load only when relevant.

PDF Generation

Slides Thief - Turn Photographed Presentation Slides into Clean PDFs

Just take photos of the slides—they are yours now.

CLI

MemoryCustodian: Durable Project Memory for Coding Agents

Memory can grow; context must stay small.

EDA

Social inequality and cultural factors impact the awareness and reaction during the cryptic transmission period of the pandemic

PNAS Nexus

COVID-19 Data Exploration (WIP)

Exploration and visualization of COVID-19-related data

Cashbox Magazine Song Chart Analysis

A simple example of a data analysis pipeline

Think Strategically and Get Jobs - H1B Visa Analysis

Helping international data science students strategize their job search

What makes a trending video?

An analysis of trending YouTube videos

COVID-19

Social inequality and cultural factors impact the awareness and reaction during the cryptic transmission period of the pandemic

PNAS Nexus

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

COVID-19 Data Exploration (WIP)

Exploration and visualization of COVID-19-related data

Data Science

Social inequality and cultural factors impact the awareness and reaction during the cryptic transmission period of the pandemic

PNAS Nexus

Think Strategically and Get Jobs - H1B Visa Analysis

Helping international data science students strategize their job search

As data scientists, what should we do?

Data science application analysis, using product launch processes as an example

Paper

Social inequality and cultural factors impact the awareness and reaction during the cryptic transmission period of the pandemic

PNAS Nexus

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

Machine Learning

Characterizing browser-based medical imaging AI with serverless edge computing: towards addressing clinical data security constraints

SPIE Medical Imaging, 2023

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

Deep Learning

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

Darknet to Keras - Model Transformation

New version based on TensorFlow 2.0

Edge Computing

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

Masked Face

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

Serverless

WearMask: Fast in-browser face mask detection with serverless edge computing for COVID-19

Electronic Imaging

Python

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Darknet to Keras - Model Transformation

New version based on TensorFlow 2.0

How to Get Information from an IP Address?

Note - geoip2 module introduction

What makes a trending video?

An analysis of trending YouTube videos

Text Analysis

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

Cashbox Magazine Song Chart Analysis

A simple example of a data analysis pipeline

Anomaly Detection

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

K-Means

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

PySpark

K-Means-Based Anomalous Email Detection in PySpark

Anomaly detection for emails based on MinHash and K-Means, implemented with PySpark and Colab

C++

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

Model Deployment

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

NCNN

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

WebAssembly

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Object Detection

WearMask - Face Mask Detection Project Details

Low-cost solution for small businesses and public places

Super-Fast In-Browser Face Mask Detection

Open the web page, and you have it!

Tutorial

Tutorial for Compiling NCNN with WASM

The second step to deploying a deep learning model in the browser

Tutorial for Compiling the NCNN Library

The first step to deploying a deep learning model

How to Get Information from an IP Address?

Note - geoip2 module introduction

R

COVID-19 Data Exploration (WIP)

Exploration and visualization of COVID-19-related data

LAPOP Data Dashboard

A dashboard prototype built with Shiny R

Cashbox Magazine Song Chart Analysis

A simple example of a data analysis pipeline

Think Strategically and Get Jobs - H1B Visa Analysis

Helping international data science students strategize their job search

Power and Sample Size Calculations for Correlational Studies

Probability and Statistical Inference - 10

Central Limit Theorem - Approximation

Probability and Statistical Inference - 09

What makes a trending video?

An analysis of trending YouTube videos

Simulation Method Comparison

Probability and Statistical Inference - 08

Coverage Probability of MLE

Probability and Statistical Inference - 07

Which quantiles of a continuous distribution can one estimate with more precision?

Probability and Statistical Inference - 05

If home-field advantage exists, how much of an impact does it have on winning the World Series?

Probability and Statistical Inference - 04

How often does the better team win the World Series?

Probability and Statistical Inference - 03

What’s the difference between Absolute Error and Relative Error?

Probability and Statistical Inference - 02

Roulette Simulation

Probability and Statistical Inference - 01

Visualization

COVID-19 Data Exploration (WIP)

Exploration and visualization of COVID-19-related data

LAPOP Data Dashboard

A dashboard prototype built with Shiny R

Visualization Cheat Sheet

A quick reference for efficient data visualization

Dashboard

LAPOP Data Dashboard

A dashboard prototype built with Shiny R

Career

Think Strategically and Get Jobs - H1B Visa Analysis

Helping international data science students strategize their job search

As data scientists, what should we do?

Data science application analysis, using product launch processes as an example

Talk

As data scientists, what should we do?

Data science application analysis, using product launch processes as an example

Simulation

Power and Sample Size Calculations for Correlational Studies

Probability and Statistical Inference - 10

Central Limit Theorem - Approximation

Probability and Statistical Inference - 09

Simulation Method Comparison

Probability and Statistical Inference - 08

Coverage Probability of MLE

Probability and Statistical Inference - 07

Which quantiles of a continuous distribution can one estimate with more precision?

Probability and Statistical Inference - 05

If home-field advantage exists, how much of an impact does it have on winning the World Series?

Probability and Statistical Inference - 04

How often does the better team win the World Series?

Probability and Statistical Inference - 03

What’s the difference between Absolute Error and Relative Error?

Probability and Statistical Inference - 02

Roulette Simulation

Probability and Statistical Inference - 01

Statistics

Power and Sample Size Calculations for Correlational Studies

Probability and Statistical Inference - 10

Central Limit Theorem - Approximation

Probability and Statistical Inference - 09

Simulation Method Comparison

Probability and Statistical Inference - 08

Coverage Probability of MLE

Probability and Statistical Inference - 07

Which quantiles of a continuous distribution can one estimate with more precision?

Probability and Statistical Inference - 05

If home-field advantage exists, how much of an impact does it have on winning the World Series?

Probability and Statistical Inference - 04

How often does the better team win the World Series?

Probability and Statistical Inference - 03

What’s the difference between Absolute Error and Relative Error?

Probability and Statistical Inference - 02

Roulette Simulation

Probability and Statistical Inference - 01

Cheat Sheet

Visualization Cheat Sheet

A quick reference for efficient data visualization

Keynote

Visualization Cheat Sheet

A quick reference for efficient data visualization