Software solutions for dirty jobs

---

I've had a lot of jobs.

Marine. Diplomatic security in Iraq and Afghanistan. Oil field mechanic at a rathole shop. Lease operator driving around West Texas checking pumping units. Pipeline operator.

None of those jobs had anything to do with software. But all of them had one thing in common: terrible tools and way too much time spent on paperwork.

When I was working as a pumper, I was basically a human computer. Drive out to the middle of nowhere. Check the gauges. Measure the tanks. Write it down. Drive to the next site. Repeat. Then go home and spend another hour punching numbers into a spreadsheet.

That's what got me into coding. Not some grand plan to become a developer. I just wanted to automate the parts of my job that were driving me crazy.

---

Muddy Boots Code

I taught myself Python while working full time and going to school. My first real win was a script that pulled data out of a 1,600-page PDF and dumped it into Excel. That task used to take someone two weeks by hand. My script did it in minutes.

That's when I knew what I wanted to build.

I was spending a lot of time around petroleum Landmen back then. If you've never met a Landman, they're the people who do all the paperwork so oil companies can drill wells. Title research, lease negotiations, right-of-way agreements—mountains of documents, most of it still managed with spreadsheets and filing cabinets.

I've sat through enough dinner conversations about terrible industry software to last a lifetime. Everyone complained. Nobody did anything about it.

So I started Muddy Boots Code with a simple idea: software solutions for dirty jobs.

Not another social media app. Not chasing whatever framework is hot this week. Just practical tools for industries that most developers are never exposed to.

I built a mineral rights app using Neo4j to untangle the mess that happens when land ownership passes through four generations of a Texas family. I spent two years at MineralSoft building SaaS for Landmen. Side projects scraping production data from state databases.

Oil and gas. That's my wheelhouse.

---

The Call

Earlier this year, I got a message from my former boss at MineralSoft. A group of drilling engineers at Trade Craft needed someone to build a wellbore diagram tool.

He recommended me. That's how this industry works—relationships and reputation. Someone puts their name behind you, doors open.

When I talked to the Trade Craft guys, I got it immediately.

A wellbore diagram is basically a schematic of a well. It shows the hole, the casing, the cement, perforations, any junk that's stuck downhole. Engineers use them for everything—planning jobs, communicating with the rig crew, documenting what's actually in the ground.

And the tools for making them? About what you'd expect.

---

The Problem

Most engineers either hand-draw their diagrams or fight with Excel. The fancy option is a desktop app called Well Shadow from PetroCode. It's fine. Better than Excel, at least.

But it's still one engineer, one machine, one file.

Well Shadow has version history—you can look back and see what changed. But there's no version control. You can't branch off a diagram to play with a proposed workover without messing up the original. You can't tag a baseline and compare it to what you're planning. You can't see the lineage of how a well diagram evolved over three years and twelve jobs.

Collaboration? Hope you like emailing PDFs.

The drilling engineers at Trade Craft had been dealing with this for years. They wanted something better.

---

What We're Building

They wanted Git for wellbore diagrams. That's the simplest way to explain Well Wise.

Branch when you need to try something. Tag when you've got a known-good baseline. Compare proposed changes against current state. Track who changed what. Roll back when something goes sideways.

Cloud-native, so the whole team works off the same source of truth. Multi-tenant, so different companies get real data isolation. And we plugged into California's public wells database—241,000+ wells from WellSTAR—so you can search and import actual well data instead of starting from a blank screen.

We built it on AWS with Lambda, DynamoDB, and OpenSearch. Serverless from the ground up. The version control isn't bolted on—it's baked into the data model with branching, tagging, and full lineage tracking.

It's been a good project. The kind where you go to bed thinking about the problem and wake up wanting to get back to it.

---

What's Coming

This is the first post in a series where I'll dig into how we built Well Wise. Real code, real problems, real solutions.

Here's what I'm planning to cover:

• Building SaaS on AWS with SST v3 – Infrastructure-as-code, multi-tenant isolation, real-time search
• React + Konva for Technical Diagrams – 2D canvas rendering and the geometry headaches that come with it
• The Dual Coordinate System – How we made canvas labels work across different screen sizes
• Refactoring for Scale – Centralizing state management without breaking production
• Multi-Tenant Architecture with Cognito – Organizations, roles, invitations
• Integrating 241,943 Public Wells – Syncing a state database and making it searchable

If you're building visualization tools, SaaS for niche industries, or just curious how serverless holds up under real workloads—stick around.

---

Well Wise

Well Wise is a professional wellbore data management and visualization platform built for drilling engineers who need more than spreadsheets and desktop apps. Upload, process, and visualize wellbore data with advanced diagramming tools—all in the browser, all synced across your team.

We're currently in private beta with Trade Craft Engineers. Public launch is coming in 2025, along with expanded state database integrations beyond California.

Want early access or have questions? Reach out on LinkedIn or drop me a line.

---

Michael Porter is the founder of Muddy Boots Code and a Graph Engineer at Graphable. He builds software for industries off the beaten path. Find him on LinkedIn, Medium, or GitHub.