will turn out to be our digital assistants, serving to us navigate the complexities of the fashionable world. They’ll make our lives simpler and extra environment friendly.” Inspiring and fully unbiased assertion from somebody who already invested billions on this new expertise.
The hype is actual for AI brokers, and billions are pouring in to construct fashions that can make us extra productive and extra artistic. Exhausting to disagree once I fortunately take pleasure in my morning espresso whereas Cursor is coding my unit checks. But, asking individuals in my community how they use AI of their day-to-day, their solutions typically point out anecdotal use circumstances, wherever from “I take advantage of it to inform bedtime tales to my son” (I assume that will not even be a use case should you had extra creativeness) to “I take advantage of it to optimize my schedule” (Movement AI, please cease focusing on me for the love of god).
As a Knowledge Scientist, my thoughts goes forwards and backwards between two conclusions. The FOMO a part of me that doesn’t need to be late to the Robotic revolution celebration, and the cynical one which thinks that there’s nonetheless a protracted strategy to go earlier than synthetic intelligence truly turns into clever. To search out out which aspect of my schizophrenic persona I ought to wager on, I’m going to make use of a easy but highly effective framework: reviewing all of the tasks I’ve labored on because the starting of my profession and assessing how 2025 state-of-the-art AI fashions may have helped.
As we speak, we return to 2018. I’m a candid summer season intern at one of the disruptive startups in America: Hire the Runway.
What the Venture was about
The Hire the Runway achievement heart in Secaucus, NJ, was once the largest dry cleansing facility in the US.
Within the Summer season 2018, as an Operations Analyst intern, I used to be given a fairly exhausting downside to consider: on a regular basis, the achievement heart was receiving 1000’s of items again from throughout the nation. All of the objects needed to be first inspected, then would undergo a radical cleansing course of, earlier than being dried or receiving some particular remedies. This could possibly be:
- Recognizing if the garment was stained through the rental
- Urgent if it was too wrinkled and needed to be ironed
- Repairing if it had been broken
Most of those duties had been finished manually by totally different departments, and required specialised staff to be out there as quickly as the primary batch of items had been reaching their division. With the ability to predict days forward what quantity of items must be processed (and when) was essential for the achievement heart planning squad, with the intention to be sure that each operations group can be staffed appropriately.
The complexity of the movement made it even trickier. It was not solely about predicting the inbound quantity, but additionally assessing what a part of this inbound quantity would require particular remedies, the place and when bottlenecks may seem, and understanding how the work finished at one division would influence the opposite departments.
The 2018 Resolution
At this level you could surprise: given the complexity and the stakes of the venture, why was it within the fingers of a younger inexperienced intern? To be truthful, throughout my 10-week summer season internship, I solely scratched the floor and wrote an insanely sophisticated Pyomo script that was later refined by a extra senior Knowledge Scientist, who spent two years on this venture alone.
However as you may think about, the answer was this big optimization mannequin taking as an enter the inbound quantity forecast for every single day of the week, the common UPH (items per hour, i.e the variety of items that may be processed in an hour) at every division, and a few assumptions on the proportions of items that will require particular remedies. The primary constraints had been on the timing and regularity of the shifts, and the variety of full time contracts. The mannequin would then output an optimized labor planning for the week.
How AI may have helped
Let’s re-clarify issues first: you’ll not see phrases like “AI-enthusiast” or “LLM believer” in my LinkedIn bio. I’m fairly skeptical that AI will magically resolve all our issues, however I’m all for seeing if with in the present day’s expertise, one other strategy can be attainable.
As a result of our strategy was, you may say, fairly old-fashioned, and required months and months of refinements and testing.
The primary restrict is the static side of the answer. If one thing surprising occurs through the week (e.g a snow storm that paralyzes the logistics in some elements of the nation, delaying a number of the inbound quantity), a variety of assumptions of the mannequin must be modified, and its outcomes have gotten out of date.
This can be a resolution that requires information scientists to go deep into the weeds, as an alternative of counting on an out-of-the-box framework, to depend on a variety of assumptions and to spend time sustaining and updating these assumptions.
May AI give you a very totally different strategy? No.
For this explicit downside, you clearly want an optimization mannequin, and I’m but to examine an LLM with the ability to deal with a mannequin with such complexity. One may suggest a framework with an AI agent performing as a Common Supervisor, and counting on sub-agents to deal with the planning of every division. However that framework would nonetheless require brokers to have instruments that enable them to unravel a fancy optimization mannequin, and the sub-agents would want to speak because the scenario of 1 division can have an effect on all of the others.
May AI considerably improve the “human-generated” resolution? Attainable.
It’s at this level fairly apparent to me that LLMs wouldn’t make the issue trivial, however they may assist enhance the answer in a number of areas:
- Initially, they may assist with reporting and resolution making. The output of the optimization mannequin may need a enterprise sense, however making a choice out of it could be exhausting for somebody with no sturdy understanding of linear programming. An LLM may assist interpret the outcomes and counsel concrete enterprise choices.
- Secondly, an LLM may assist react sooner to sure surprising conditions. It may for instance summarize info on occasions that would have an effect on the Operations, corresponding to dangerous climate in some elements of the nation or different points with suppliers, and as such, suggest when to rerun the planning mannequin. That’s assuming it has entry to good high quality information about these exterior occasions.
- Lastly, it’s attainable AI may have additionally helped with making actual time changes to the planning. As an illustration, it’s sometimes predictable based mostly on the garment traits whether or not they would require particular care (e.g a cotton shirt will at all times must be ironed manually). Having a VLM scanning each garment on the receiving station may assist downstream departments perceive how a lot quantity they need to anticipate hours prematurely.
May AI allow Knowledge Scientists to take care of and replace the mannequin? Sure!
It’s actually exhausting to disclaim that with instruments like Copilot or Cursor coding and sustaining this mannequin would have been simpler. I’d not have blindly requested Claude to code each constraint of the Linear Program from scratch, however with AI code editors being smarter than ever, modifying and testing particular constraints (and catching human errors!) can be simpler.
My conclusion is that an LLM in 2018 wouldn’t have trivialized the venture, though it may have enhanced the ultimate resolution. However it’s not unattainable to consider that just a few years (months?) from now, brokers with enhanced reasoning capabilities can be subtle sufficient to start out cracking these kind of issues. Within the meantime, whereas AI may velocity up mannequin iterations and changes, the human judgment on the core stays irreplaceable. This serves as a beneficial reminder that being a Knowledge Scientist isn’t nearly fixing mathematical or pc science issues—it’s about designing sensible options that meet evolving, typically ambiguous and never so effectively outlined real-world constraints.
Article 100% human generated
