How I Built a Local PDF PII Redaction Tool Before Uploading Documents to AI

I wanted to use AI to analyze a sensitive PDF, but I did not want to upload the original document with personal information intact. The document was also large, around 60 pages, with names, identifiers, contact details, and other PII repeated many times. Manually finding and redacting every occurrence would have been tedious and easy to get wrong.

My requirements were:

• Run completely locally on Windows.

• Open and process a large PDF, around 60 pages.

• Automatically detect likely PII across the document.

• Find repeated names, emails, phone numbers, dates, IDs, and similar fields.

• Let me review the detections before applying them.

• Keep useful analytical fields such as age, premium, cash value, surrender value, MEC information, loan provisions, and policy values.

• Export a sanitized PDF that I could then upload to ChatGPT, Claude, Gemini, or another AI tool.

The goal was:

Original PDF -> Local PII Detection -> Review -> Redact -> Sanitized PDF -> AI

Finding the Right Tool

I looked at OpenRedact, PDF24, PDFgear, PDF-XChange, and a few other options.

Most traditional PDF tools work well when you already know exactly what needs to be removed. My requirement was slightly different: I wanted the software to help identify what might be PII first.

I eventually found:

PII GUI

https://github.com/sophia486/pii-gui

It is an open-source desktop application built with Tauri, React/TypeScript, Rust, and ONNX-based local detection.

Why I Built It From Source

The downloaded Windows installer triggered Microsoft Defender SmartScreen:

Windows protected your PC
Publisher: Unknown publisher

The executable also had no Digital Signatures tab in Windows Properties.

That does not automatically make an application unsafe, but since I planned to use it for sensitive documents, I decided to build it directly from the GitHub source instead.

Clone the Project in VS Code

I used VS Code because the project uses TypeScript, React, Rust, and Tauri.

Press:

Ctrl + Shift + P

Choose:

Git: Clone

Repository:

https://github.com/sophia486/pii-gui.git

My local folder was:

D:\ch-projects\pii-gui\pii-gui

Install the Required Tools

Check what is already installed:

node --version
pnpm --version
rustc --version
cargo --version

I already had Node.js.

Install pnpm

npm install -g pnpm

Verify:

pnpm --version

Install Rust and Cargo

Install Rust from:

https://rustup.rs/

Then verify:

rustc --version
cargo --version

If VS Code cannot find them:

where.exe rustc
where.exe cargo

Rust normally installs under:

C:\Users\<username>\.cargo\bin

Install Visual Studio C++ Build Tools

Open Visual Studio Installer and install or update:

Build Tools 2022 -> Modify -> Desktop development with C++

Make sure the latest available versions of these are installed:

• MSVC v143 C++ build tools

• Windows SDK

• C++ CMake tools

This became important because ONNX Runtime depends on native C++ libraries.

Install Project Dependencies

Go to the Tauri directory:

cd D:\ch-projects\pii-gui\pii-gui\tauri

Then:

pnpm install

Run the development version:

pnpm tauri dev

Fixing the ONNX Runtime Linker Error

My first build produced a large number of linker errors such as:

LNK2001: unresolved external symbol __std_find_last_of_trivial_pos_1

and:

__std_search_1
__std_remove_2

The problem was not React, Rust, or Tauri code.

The ort-sys dependency was using a statically linked ONNX Runtime built with a newer MSVC C++ Standard Library than my installed Visual Studio 2022 Build Tools 17.10 provided.

The fix was to update the MSVC v143 C++ toolset and Windows SDK using Visual Studio Installer.

Then I cleaned the previous Rust build:

cd D:\ch-projects\pii-gui\pii-gui\tauri\src-tauri
cargo clean
cd ..
pnpm tauri dev

Port 1420 Already in Use

At one point Vite reported:

Port 1420 already in use

Find the process:

netstat -ano | findstr :1420

Then stop it:

taskkill /PID <PID> /F

Run again:

pnpm tauri dev

Opening a PDF

The GUI did not make the PDF-open option obvious.

The shortcut is:

Ctrl + O

That opens the Windows file picker.

You can then select the PDF and load it into PII GUI.

For my use case, this was important because the document was about 60 pages long and contained many repeated PII entries.

Instead of manually working page by page, the workflow became:

60-page PDF -> Detect PII Across Document -> Review Matches -> Apply Redactions -> Export

Reviewing PII

PII GUI can detect categories such as:

PrivatePerson
PrivateEmail
PrivatePhone
PrivateDate
PrivateUrl
Secret
UserId

The key part is that the results can be reviewed before redaction.

That matters because not every detected value should be removed. In an insurance illustration, for example, I may want to preserve:

• age

• underwriting class

• premium

• face amount

• cash value

• surrender value

• MEC values

• loan information

• policy charges

while removing names, addresses, phone numbers, and identifiers.

Build the Windows Installer

After confirming the app worked with:

pnpm tauri dev

I created the release build.

Tauri generated both an NSIS .exe installer and an MSI installer under:

D:\ch-projects\pii-gui\pii-gui\tauri\src-tauri\target\release\bundle

The NSIS installer was under:

...\bundle\nsis

The MSI installer was under:

...\bundle\msi

I verified them using:

Get-ChildItem `
'D:\ch-projects\pii-gui\pii-gui\tauri\src-tauri\target\release\bundle\nsis', `
'D:\ch-projects\pii-gui\pii-gui\tauri\src-tauri\target\release\bundle\msi' |
Select-Object FullName,Length,LastWriteTime

Tauri Signing Warning

The release command returned exit code 1 because updater signing was enabled but:

TAURI_SIGNING_PRIVATE_KEY

was not configured.

The message was:

A public key has been found, but no private key

The important point was that the installers had already been generated successfully. The failure was related to updater signing, not the application build itself.

Install the Locally Built Application

The generated installer was:

D:\ch-projects\pii-gui\pii-gui\tauri\src-tauri\target\release\bundle\nsis\PII GUI_0.1.2_x64-setup.exe

After running it, PII GUI was installed like a normal Windows application and could be launched from the Start menu.

Final Workflow

My everyday process is now:

Launch PII GUI -> Ctrl + O -> Open PDF -> Detect PII -> Review -> Redact -> Export Sanitized PDF -> Upload to AI

For the blog screenshots, I only use pages from the final sanitized PDF. I do not publish the original before-image because it contains real PII.

Bottom Line

The original problem was simple: I had a large, approximately 60-page PDF with many repeated personal-information fields, and I wanted to use AI to analyze it without sending the original PII.

Building PII GUI locally gave me a reusable workflow:

Sensitive PDF -> Local PII Detection -> Human Review -> Redacted PDF -> AI

The most important lesson is straightforward:

Sanitize locally first. Use AI second. And here is a sample example of one of the pages of a 60-page document.

How I Downloaded a 2-Hour YouTube School Convocation Video Using yt-dlp

 

I had a 2-hour school convocation video on YouTube that I wanted to save for my own future reference. I first tried using a GUI downloader, but it kept spinning and never actually downloaded the video. The working solution was to use yt-dlp, plus a JavaScript runtime and FFmpeg support.

This guide documents the exact process that worked on Windows.

---

1. Install yt-dlp

Download yt-dlp from the official GitHub release page:

https://github.com/yt-dlp/yt-dlp/releases

After downloading, place yt-dlp.exe in a folder such as:

D:\ch-software

Then open Command Prompt and go to that folder:

cd /d D:\ch-software

Test that it works:

yt-dlp --version

---

2. Install Deno for YouTube JavaScript Challenges

When I first ran yt-dlp, I saw a warning like this:

No supported JavaScript runtime could be found.
YouTube extraction without a JS runtime has been deprecated.

To fix this, I installed Deno using Windows Package Manager:

winget install DenoLand.Deno

Then I closed Command Prompt, reopened it, and checked:

deno --version

If winget is not recognized, install or update App Installer from the Microsoft Store.

---

3. Install FFmpeg

FFmpeg is important because YouTube livestream replays may download as HLS fragments. Without FFmpeg, the file may have timestamp issues or slight audio/video sync problems.

Install FFmpeg:

winget install Gyan.FFmpeg

Then close and reopen Command Prompt.

Check:

ffmpeg -version

---

4. Check Available Video Formats

Before downloading, I checked what quality options were actually available:

yt-dlp -F "https://www.youtube.com/watch?v=yZFp6eo91rM"

The output showed this:

91 mp4 256x144     15
92 mp4 426x240     30
93 mp4 640x360     30
94 mp4 854x480     30
95 mp4 1280x720    30

This confirmed that the best available version was 720p, not 1080p. If YouTube does not provide a 1080p stream, yt-dlp cannot create one.

---

5. Create a Custom Download Folder

I wanted to save the video into:

D:\my videos

So I created the folder:

mkdir "D:\my videos"

---

6. Best Working Download Command

The final command that worked was:

yt-dlp --fixup force --remux-video mp4 -o "D:\my videos\%(title)s_%(height)sp.%(ext)s" "https://www.youtube.com/watch?v=yZFp6eo91rM"

What this command does

Command part	Meaning
yt-dlp	Runs the downloader
no -f option	Lets yt-dlp choose the best available format automatically
--fixup force	Forces post-download fixes for timestamp/container issues
--remux-video mp4	Repackages the video into a clean MP4 without re-encoding
-o "D:\my videos\%(title)s_%(height)sp.%(ext)s"	Saves the file to the custom folder and includes the resolution in the filename
YouTube URL	The video to download

---

7. Important Note About -f best

I originally tried using:

-f "best"

But yt-dlp gave this warning:

"-f best" selects the best pre-merged format which is often not the best option.
To let yt-dlp download and merge the best available formats, simply do not pass any format selection.

So the better approach is to omit -f entirely and let yt-dlp pick the best available video/audio combination.

---

8. Understanding the Fragment Download

During the download, I saw:

[hlsnative] Total fragments: 1216

This is normal for a YouTube livestream replay. YouTube serves the video as many small HLS fragments. yt-dlp downloads all the fragments and assembles them into one file.

---

9. If Audio Is Slightly Out of Sync

After one earlier download, the file had a slight audio/video sync issue. The fix is to remux the MP4 using FFmpeg:

ffmpeg -i "input.mp4" -c copy "fixed_remux.mp4"

This does not reduce quality. It simply rebuilds the container.

If audio is still slightly off, shift the audio manually.

If audio is early

ffmpeg -i "input.mp4" -itsoffset 0.25 -i "input.mp4" -map 0:v -map 1:a -c copy "fixed_audio_delayed.mp4"

If audio is late

ffmpeg -i "input.mp4" -itsoffset -0.25 -i "input.mp4" -map 0:v -map 1:a -c copy "fixed_audio_advanced.mp4"

A small issue is usually around 0.1 to 0.4 seconds.

---

Final Command to Keep for Future Use

For the same type of YouTube school convocation video, this is the command I would keep:

mkdir "D:\my videos"
yt-dlp --fixup force --remux-video mp4 -o "D:\my videos\%(title)s_%(height)sp.%(ext)s" "YOUTUBE_URL_HERE"

Example:

mkdir "D:\my videos"
yt-dlp --fixup force --remux-video mp4 -o "D:\my videos\%(title)s_%(height)sp.%(ext)s" "https://www.youtube.com/watch?v=yZFp6eo91rM"

---

Bottom Line

The GUI downloader did not work, but yt-dlp worked once the proper supporting tools were installed. The key pieces were:

1. yt-dlp for downloading.

2. Deno for YouTube JavaScript challenge handling.

3. FFmpeg for clean MP4 remuxing and timestamp fixes.

4. No -f best; let yt-dlp choose the best available format automatically.

5. Use --fixup force and --remux-video mp4 for cleaner final output.

Access Boomi Insights via Postman: Ready-to-Use Collection

Following our previous guide on retrieving your Boomi Public API Key, this post shows how to set up a Postman collection and environment to query telemetry metrics from Boomi Insights.

Using Postman allows you to test APIs, explore available widgets, and integrate metrics into dashboards or automation scripts.

---

Step 1: Import the Collection and Environment

1. Download the files:

   • 📑 Boomi_Insights_Collection.postman_collection.json

   • 🌍 Boomi_Insights_Env.postman_environment.json

2. Open Postman, then import both files.

3. Select the imported environment in the top-right corner.

---

Step 2: Configure Environment Variables

The environment uses three key variables:

Variable	Description
API_KEY	Your Boomi Public API Key 🔑
STACK	Your runtime stack name 🏗️
REGION_GATEWAY	Your Boomi region gateway 🌐 (e.g., us-east-prod-gateway)

💡 Tip: Updating these once applies them to all requests in the collection.

---

Step 3: Example Requests in the Collection

Name	Method	Description
CPU Metrics	GET	Fetches CPU utilization percentage for the stack
Memory Metrics	GET	Returns RAM usage percentage
Worker Count	GET	Number of active workers in the stack
Worker Execution Count	GET	Total executions by workers
List Widgets	GET	Retrieve all available telemetry widgets

Each request uses environment variables to keep URLs and API keys generic and reusable. Sample Request Structure All requests follow the same pattern: Endpoint URL: Uses REGION_GATEWAY and STACK variables Header: x-api-key uses API_KEY variable ⚡ Note: The {{VARIABLE}} syntax pulls values from your Postman environment, making it easy to switch stacks or regions. Step 4: Benefits of Using This Collection Test multiple endpoints quickly without manual URL edits Easily switch between runtime stacks or regions Standardize metrics collection across your team Serve as a foundation for dashboards or automated reporting

Using Boomi API Insights: A Step-by-Step Guide

Boomi provides API Insights, allowing you to monitor your integration runtime environments programmatically. From CPU and memory usage to worker counts and execution metrics, you can access telemetry data through the Insights API and integrate it into dashboards or monitoring tools.

This guide is split into two parts:

1. Retrieving your Boomi Public API Key

2. Accessing telemetry data using Postman

---

Part 1: Retrieving the Boomi Insights Public API Key

Before querying Boomi Insights APIs, you need an API Token and a Public API Key.

1. Generate an API Token

1. Log in to your Boomi account.

2. Go to Settings → User Profile.

3. In the left pane, click Platform API Tokens.

4. Generate a new token linked to your user account.

2. Access Boomi Insights

1. Navigate to Boomi Insights (region-specific URL).

2. Enter the following when prompted:

   • Account ID (where the token was created)

   • Username (associated with the token)

   • API Token

3. Get the Public API Key

1. Go to Managed Cloud Services → Runtime Dashboard.

2. Select your runtime stack.

3. Click Settings → Public API Settings.

4. Click Get API Key to retrieve it.

⚠️ Note: You must have the Boomi Insights Admin role to access the Public API Key.

---

Part 2: Accessing Insights Data Using Postman

Once you have the Public API Key, you can query telemetry metrics. Use a generic Postman collection with environment variables for easy configuration.

Generic API Endpoints

Metric	Endpoint	Description
CPU Usage	/telemetry?widget=cpu.utilisation.%25&stackname=YourStack	Shows CPU utilization
Memory Usage	/telemetry?widget=ram.used.%25&stackname=YourStack	Tracks memory consumption
Worker Count	/telemetry?widget=boomi.aw.count&stackname=YourStack	Returns the number of active workers
Worker Execution Count	/telemetry?widget=boomi.awExec.count&stackname=YourStack	Shows worker executions

🧘‍♂️ Monk Fasting: My 36-Hour Experiment with Hunger, Heritage, and Hot Water

Fasting has always fascinated me. Whether it's Karthika Masa Naktalu (evening-only meals), Ekadasi Ekabhuktam (twice a month), or my weekly Thursday fasting ritual, I’ve danced with hunger in different forms. And then came Monk Fasting—where you skip food for a full 36 hours. The idea was floated to me by one of the Educational Units (EUs) of WhatsApp University, and the benefits sounded like something out of a wellness brochure: detoxified digestion, sharpened focus, fat loss, and autophagy (that sci-fi-sounding cell-cleaning process). Despite years of fasting practice, sleeping with a completely empty stomach? That I could never dare.

☕The Hardest Part: Saying No to South Indian Coffee

Let’s talk priorities. I’m a coffee person—and not just your standard drip guy. My morning ritual involves mixing coffee powder with chicory root (3:1), brewing it in a moka pot, then blending that magic with freshly boiled whole milk and a generous splash of sugar. That, my friend, is South Indian coffee. After that comes mid-morning Indian tea, followed by an evening tea. And yes, the prep is a ceremony in itself. For this Monk Fast, I ditched the whole routine. No coffee. No tea. Just plain hot water, all day. Energy-wise, I was fine. But by evening, caffeine withdrawal kicked in and I got drowsy. At 10 PM—a whole two hours earlier than my usual schedule of watching the date change—I crashed. Slept like a baby until morning.

📆 The Timeline

• Started fasting: Wednesday 10 PM 
•  Broke fast: Friday 10 AM 
• Total intake: Hot water. And a ridiculous amount of it.

📉 Results (and a Marital Audit)

I weighed myself before and after: five pounds down. My wife, naturally skeptical, supervised the weigh-in like a hawk. Verdict? Certified by the Ministry of Marital Surveillance.

🤔 What I Learned

•  ✅ Weight loss: Probably temporary, but satisfying. 
•  🛑 Digestive break: Felt like my stomach got a spa treatment. 
•  😴 Drowsiness: Could’ve been the lack of caffeine more than the fasting.

🔄 Will I Do It Again?

Absolutely. It gave me more confidence around extended fasting and more respect for my usual caffeine reliance. If you're even half as crazy as I am, take it slow. Choose a day when you don't need to be an unpaid Uber driver or run marathons—and make sure your family’s nearby to make sure you are still alive.

Powershell hacks: How can I run powershell if I am not an admin

Dealing with admin restrictions on my laptop has become quite a hassle. My company won't allow me to have admin privileges, which means even running simple PowerShell commands requires me to ask for admin rights. This has added an extra layer of frustration to my workflow. Why do I need PowerShell, you might ask, when Python can solve many problems with scripts? Well, the catch is that even installing Python and importing libraries requires admin rights—leaving me stuck in a loop of limitations. After a bit of exploration, I found a way to work around this. Here's a simple trick:

powershell.exe -ExecutionPolicy bypass

Docker Diaries: Docker Pull Request for Ubuntu got Denied

Trying to install ubuntu image and executed the following command using Windows Power Shell. Apologies for not taking the screen capture. It looks something like this:

docker pull ubuntu default is latest Trying to pull ubuntu pull access denied for ubuntu does not exist or may

require 'docker login': denied: requested access to the resource is denied


So I exited from the Power Shell and logged in again to Docker. Then used docker pull command voila!! it worked. This time I took the screen capture of the output though :)

Windows PowerShell Copyright (C) Microsoft Corporation. All rights reserved. PS C:\Windows\system32> docker login Authenticating with existing credentials... Login Succeeded PS C:\Windows\system32> docker pull ubuntu Using default tag: latest latest: Pulling from library/ubuntu d51af753c3d3: Pull complete fc878cd0a91c: Pull complete 6154df8ff988: Pull complete fee5db0ff82f: Pull complete Digest: sha256:747d2dbbaaee995098c9792d99bd333c6783ce56150d1b11e333bbceed5c54d7 Status: Downloaded newer image for ubuntu:latest docker.io/library/ubuntu:latest PS C:\Windows\system32>

AMQ4036 Error connecting to queue manager on Linux using MQ Explorer or RFHUTILC

Issue: MQ Explorer fails to connect to a Queue manager located on Linux server. Fails with error AMQ4036 : Access not permitted. You are not authorized to perform this operation

Setup:

User has two IDs.
• Linux user ID: j1234d
• Windows ID: mydomain\jdove

Queue Manager (QM1) is on Linux machine and the Linux ID “j1234d” has access to the queue manager.
MQ Explorer is on Windows machine where the user is logged on with Windows ID “mydomain\jdove”.
When user tries to connect to QM1 from Windows machine using MQ Explorer, he/she gets the following error.

Resolution:

MQ Explorer makes a client connection and the credentials passed to the QM1 on Linux server would be “mydomain\jdove”. Since the ID “mydomain\jdove” is not defined on the Linux server, it fails to get authenticated.

Simple way to resolve:

1. Create ID “j1234d” on your Windows machine and use MQ Explorer to connect to the QM1 on Linux Server

However above fix would need you to log off current Windows session and sign in as “j1234d”. This is a pain.
Instead, do the following:

Create ID “j1234d” on your Windows machine and use MQ Explorer to connect to the QM1 on Linux Server
Create a Windows batch script as shown :

set APPDATA=

runas /env /user: "strmqcfg -d"

So for the above example it would be:

set APPDATA=D:\mqeclipse\mqcm5dev

runas /env /user:j1234d "strmqcfg -d"

When you run the above script, at the prompt enter the password you have set for the id you created for id ‘j1234d’

The runas  command makes the MQ Explorer to run under the local id which is the equivalent of the Linux id. When we run the MQ Explorer process under the id , the credentials will be passed as is ( eg:’j1234d’) and you will be able to connect to the queue manager since this ID on Linux server has access to the queue manager QM1.

The same solution can be used  if you are getting error "Not authorized" with RFHUTILC utility. The following screen shows the error.

Use the above approach to circumvent the error.
Hope this helps!!!