From the Bucket is a podcast about industrial control systems as they actually exist in the field. Not how they look in CAD. Not how they were sold. But how they behave at 3 a.m. when production is down and someone needs answers.
Hosted by controls engineer Kealy Smith, each episode explores real projects, real mistakes, and the long-term consequences of engineering decisions. From early career chaos and endless travel to burnout, leadership, and eventually starting a company, this podcast follows the full arc of a career spent making machines work in the real world.
Dead Data: What Industrial Data Can (and Can't) Do for You
This episode is a detour from the career story because I've spent a big part of this year buried in compressor data, and it's been some of the most interesting work I've done.
Industrial data collection is a spectrum, from an operator with a clipboard all the way up to full historians logging every point. In this episode I talk through what each level actually lets you do, how the questions change between industries, and how I physically pull and analyze this data in the field. Plus three real compressor stories: a pressure valve failure you could only see in the data, a bearing quietly failing across a month of trending, and using startup data to prove a system is healthy instead of hunting for what's broken.
And then there's the trap I see all the time: dead data. Collecting everything and looking at none of it until something fails. Because the value of your data isn't how much you collect.It's whether anybody's actually paying attention before they're forced to.
26 May 2026
The shop, Where Everything Works in CAD
Two years in Dallas. Sixteen-hour days. Full autonomy to make decisions, fix problems, and drive a project forward. And the you come home.
In this episode, I talk about the part of field engineering nobody really warns you about: the transition back to the shop. The pace whiplash. The committees for decisions you used to make in thirty seconds. And the moment I realized where the phrase "it worked in CAD" actually came from.
Also, we made it to ten episodes. I have some thoughts on that too.
11 May 2026
Someday I'll Teach You How to Program
There's a reputation that follows controls engineers around. Arrogant. Difficult. Convinced they're the smartest person in the room. I'm not going to tell you that reputation is completely undeserved. In this episode I talk about how the field builds that arrogance almost by design, what it actually looked like when I was inside it and couldn't see it, and why the honest answer is that it never fully goes away.
27 Apr 2026
The Last Thing Dallas Taught Me
Every big chapter ends somewhere. This is where Dallas ends for me with miles of overhead conveyor, a burned motor, and a welding hood I hadn't picked up since high school. What two years in one place actually made me.
13 Apr 2026
Your Machine, My Line
Every machine that comes into a production line has its own PLC, its own logic, and its own idea of how the world should work. As the system integrator, that's your problem to solve. In this episode, we break down the reality of integrating third-party machines into a line control system. Why fillers are the hardest machines to work with, why Ethernet doesn't actually mean access, and what it took to stop showing up to projects blind. From a bottle jam nobody could see to vendors who just said no, this is the dull work that determines whether a line actually runs.
30 Mar 2026
Temporary (But Not Really)
The filler panel caught fire. It wasn't dramatic,just enough to trip the alarm and send everyone outside to stand in 110 degree Dallas heat. That moment pretty much sums up this entire project. This episode is about bad equipment, impossible expectations, and what you learn when software can't save you.
16 Mar 2026
Ninety-Nine Percent
An experimental filler pushing 210 cartons per minute created one unexpected challenge: merging three lanes of cartons into a single line without collisions.
What looked like a simple control problem became a deep dive into conveyor physics, compounding error, and the limits of mathematical modeling.
In this episode, I tell the story of experimental filling line, the weeks spent chasing a perfect solution, and the lesson every controls engineer eventually learns: real production systems never behave exactly like the model.
2 Mar 2026
Ratchet Straps and Good Intentions
Not every successful project looks good while it’s happening.
In this episode, I tell the story of a large gantry-based X-ray inspection system that started in frustration and ended in success. The original design never really existed, the space was brutally constrained, and normal installation methods were off the table from day one.
I talk through why we started from scratch, how a distributed motion system unlocked the design, and how we ultimately assembled the machine using modular construction, furniture dollies, ratchet straps, and a willingness to accept that there were no perfect options.
This is a behind-the-scenes look at the kind of engineering that happens when reality refuses to cooperate.
Music from Pixabay Monument_Music
16 Feb 2026
Travel, Responsibility, and Becoming “The Guy”
Early career travel and responsibility often arrive together in controls engineering. In this episode, I talk about what “paying your dues” actually looked like for me, living on site, owning multiple lines too early, and learning the job under constant pressure. This is a story about growth, pride, and the hidden costs of becoming reliable.
Music from Pixabay Monument_Music
16 Feb 2026
We’ve Built This Before
Early in my career, I was handed my first real production line and told not to worry, the company had built it before. What I wasn’t given was the code, the standards, or a real starting point.
In this episode, I talk through what happens when “experience” exists only as stories, how abstraction becomes a survival mechanism, and how early habits can quietly turn into long term technical debt. From programming a blank controller in the field to learning hard lessons during first production, this episode is about momentum, responsibility, and figuring things out before you know what good actually looks like.
If you’ve ever been told “we’ve done this before” and then left to invent the structure yourself, this one will feel familiar.
Music from Pixabay Monument_Music
7 Feb 2026
The Code That Wouldn’t Go Away
In this first episode of From the Bucket, I tell the story of my very first controls project.
It was a barcode verification system that worked exactly the way it was designed. We deployed it, customers were happy, and nothing broke.
Years later, after feature creep, expansion, and a lot of well-intentioned decisions, it became a system that only a few people could support. And one of those people was me.
This episode isn’t about a dramatic failure or a clean fix. It’s about how control systems really fail. Slowly, quietly, and over time. It’s about abstraction, maintainability, and what happens when early design decisions turn into legacy systems that you have to live with.
If you’ve ever opened up a program that technically works but nobody really understands anymore, this one will probably feel familiar.
Music from Pixabay Monument_Music
Reach and audience
Public platform figures. Ratings count people who left a rating, not total listeners.
Apple Podcasts (US)
5.0 / 5
1 ratings
Contact From the Bucket
Guest appearances
Does not typically book guests
Based on episode analysis; this does not confirm that the show is currently accepting guests.
Host of From the Bucket?
Claim your podcast to manage its listing and keep your show details accurate.
Pod Engine is an independent podcast discovery and analytics service and is not affiliated with or endorsed by this podcast. Artwork and show content belong to their owners. Full legal notice.
Explore this show Podcast research with Pod Engine
