The staff scheduler took about 20 years to build.
That’s not as dramatic as it sounds. The MILP model — the math, the constraints, the logic — that came together during the Fiverr engagement described in the first post of this series. What took 20 years was accumulating enough Operations Research experience to know what the model needed to look like. The actual build, once I sat down with the problem fully understood, was fast.
The web app took a few days. The deployment took an afternoon, plus one failed attempt that taught me about gunicorn.
What’s Sitting in Excel Right Now
I have been doing Operations Research professionally for over 20 years. Cargill. Cessna Aircraft. Textron Aviation. Consulting projects through Datos Technologies. Each of those engagements left behind models — spreadsheets, mostly, built in Excel with SolverStudio, doing things that took commercial solver licenses and real expertise to set up.
Those models work. They’re just locked. You need the exact right version of Excel. You need SolverStudio installed. You need to know how to use them. They’re invisible to anyone who doesn’t have that setup, and they sit idle on a hard drive while the problems they solve continue to exist in the world.
The staff scheduler was one of them. It isn’t anymore.
Here’s what else is in the queue:
Vehicle Routing — Given a set of locations and a fleet of vehicles with capacity constraints, find the minimum-cost routes. Classic Operations Research problem. Every delivery company solves a version of this every day. (If you just need to map a delivery run rather than optimize a whole fleet, I rounded up some free multi-stop route planners.)
Warehouse Slotting — Where do you put things in a warehouse to minimize travel time? This sounds like a facilities management problem, and it is, but it’s also a combinatorial optimization problem with some non-obvious structure.
Least-Cost Formulation — Given a target specification and a set of available ingredients with costs and nutritional profiles, find the minimum-cost blend that meets the spec. This one has direct applications in food manufacturing, animal feed, and a few other industries I’ve worked in.
That’s three more models. There are others — I’m not going to list all 20-something here because I’m not sure which ones will actually get converted, and I’ve learned not to make promises about things that haven’t been built yet.
The Business Case
Let’s talk about what this is actually for.
I’m financially independent. I left the corporate world over a decade ago. The tradelines business at kindoflost.com generates passive income. I don’t need a new job, and I’m not trying to build a startup.
What I’m interested in is this: there are small and medium-sized businesses that have Operations Research problems — real scheduling problems, routing problems, formulation problems — and no access to people who can solve them. The tools that commercial OR consultants use (expensive licenses, specialized software, implementation fees) are priced for enterprise clients. The SMB market is largely underserved.
A web app that solves a staff scheduling problem for $7 a month in hosting costs changes that equation. You don’t need to hire a consultant. You don’t need a software license. You need an internet connection and an Excel file in the right format.
Whether that turns into a business is a separate question. Right now it’s a proof of concept. The staff scheduler works. It returns correct results. It’s live. Anyone can use it.
What FIRE Has to Do With This
I’ve thought about why this project happened now, after years of those models sitting untouched in Excel files.
Part of it is that the technology got better. A few years ago, building the Flask wrapper and deploying it to a production server would have required either hiring a web developer or spending weeks learning web development from scratch. The tooling available now — including AI coding assistants — compressed that timeline significantly. The web parts of this project took days, not weeks. That changes the math on whether it’s worth starting.
But the bigger factor is having the time and freedom to finish things that aren’t paying immediately. When you’re employed or billing by the hour, the calculus is straightforward: does this activity generate income right now? If not, it doesn’t happen. The Fiverr client who backed out on the original staff scheduler engagement — that model sat in a file for years because there was no immediate reason to finish it.
Being financially independent means the calculus changes. The question isn’t “does this pay right now?” It’s “is this interesting? Is it worth building? Will I be glad I built it?” The answers to those questions pointed at finishing the conversion.
So that’s what happened. And that’s what’s happening next.
The live app is at staff-scheduling.kindoflost.com. The sample workbook is available for download on that page. If you have a staff scheduling problem that looks like the one described in this series, give it a try.
Getting one of these online was its own adventure; I wrote about deploying the app for $7 a month.
Things that I use, like, and am affiliated with:
Mint Mobile offers great cell phone service for $15 flat, get $15 off using the link. Get discounted phones with service activation and no contract.
I never spend money before I check Mr Rebates or Rakuten to get cashbacks, rebates, discounts, coupons or cheaper gift cards.