Kindred Home — Online Lectures
AI on the
Factory
Floor
Not a survey course. Not a certification mill. A structured lecture series for people who work in manufacturing and want to understand what machine learning is actually doing to their industry — before someone else figures it out for them.
Who this is for
Someone already in the room
Tobias Krantz spent eleven years managing a mid-sized automotive parts plant in southern Ontario. He knew the machines, the schedules, the failure patterns. Then the company installed a predictive maintenance system and he found himself nodding along in meetings while understanding roughly 40% of what was being discussed.
That specific discomfort — knowing the domain but not the language — is what these lectures address. The content assumes you already understand production lines, quality targets, and what a bad shift looks like. It does not assume you know what a neural network is.
Participants come from plant operations, quality engineering, procurement, and production planning. What they share is proximity to AI tools that are already deployed — and a preference for understanding over delegation.
Mirela Ostrowski
Quality Systems Lead, Ontario
I had been approving AI-generated inspection reports for eight months before I understood what the model was actually flagging. The lectures filled that gap without making me feel like I was starting from zero.
Dariusz Fehér
Plant Operations Manager
The format worked well for shift schedules. I could follow a lecture during a quiet period and revisit specific parts later. It is not designed to be consumed in one sitting, which is honest about how manufacturing people actually have time.
The honest part
The concern most people carry in
Most people who look at this program have the same quiet worry: that it will be too technical to follow, or too basic to be worth the time. Both are reasonable concerns about most AI education available right now.
This series sits between the vendor pitch and the computer science degree. It explains how systems like computer vision quality inspection or demand forecasting models actually function — without requiring Python or a statistics background to follow along.
-
Is this too mathematical?
The lectures explain mechanisms, not derivations. You will understand what gradient descent does in a manufacturing context without working through the calculus that produces it.
-
Will it apply to my specific sector?
Examples draw from automotive, food processing, electronics assembly, and discrete manufacturing. The underlying patterns transfer; the vocabulary is grounded in things you already recognise.
-
What if I fall behind?
Lectures are recorded and sequential. There is no live cohort to keep pace with. You move at whatever speed your schedule allows, and the content does not expire.
What the commitment looks like
Time, structure, and what you get
Each module runs between 35 and 55 minutes. There are eight of them. Most participants work through one or two per week, which puts completion somewhere between one and two months depending on schedule density.
There is a written summary after each lecture, a short self-check with no grading attached, and access to a text-based discussion thread for questions. Nothing requires a camera or a live time slot.
Core lectures — recorded video with chapter markers and full transcripts
Reference material — written summaries, diagrams, and terminology guides per module
Discussion access — asynchronous thread, questions answered within two business days
Where this leads
The distance between watching and understanding
Right now, AI systems in manufacturing plants mostly get evaluated by the people who installed them. The people who operate around them often lack the vocabulary to push back, ask sharper questions, or spot when a model is behaving oddly.
After completing this series, that changes in a specific way: you will be able to read a system's performance report and know what the numbers are not telling you. That is a narrow skill with a disproportionate effect on how you participate in decisions.
This does not make you a data scientist. It makes you harder to talk past in a room that has one.
01
Read the system
Interpret AI-generated reports and performance dashboards without needing a technical translator in the room.
02
Ask better questions
Know what to probe when a model's output looks off — and what to do when the vendor's answer does not actually answer anything.
03
Stay in the conversation
Participate in AI procurement, implementation reviews, and performance evaluations as someone who understands the tradeoffs involved.