About

My name is Elias Willemse¹ and I’ve spent over 15 years studying how things move through cities. That started with a PhD on optimisation algorithms for waste collection logistics, continued through a senior lectureship at the University of Pretoria (where I taught 1500 undergraduates, which I’m never doing again), a research fellowship at SUTD Singapore with the Applied Complexity Group, and eventually a VC-backed product company called Waste Labs that helped clients win £250M+ in competitive waste and recycling tenders.

Eventually, I realised that the movement was the least interesting part of the problem. The months of analysis that go into deciding where to place infrastructure, how to allocate resources, which tender to bid on and what to promise: that reasoning is where the leverage is, and it’s where the tools are worst.

So now I build spatial knowledge graphs and AI tools for supply chains, compressing months of analysis into minutes. The deeper question, which I’m increasingly drawn to, is how groups make better decisions when the world moves faster than their tools. I wrote about this in When Deliberation Meets Reality, a response to ARIA’s Collective Flourishing programme.

The goal is, once again, products. A sensible career progression, if you squint.

The thread through it

The timeline reads as a sequence of jobs, but the thread is a single question that kept getting bigger. My PhD was about finding the best route for a truck, which is a surprisingly hard combinatorial optimisation problem when you’re routing along street segments with capacity constraints and intermediate facilities. I spent years on the algorithms and published the results.

At Waste Labs, the question expanded from one truck to entire collection networks designed under competitive tender pressure. Four years of that taught me that the algorithms mattered less than the ability to iterate on questions with a client in the room. The bottleneck was always the time between a question and a grounded answer, not the quality of the solver.

That experience is what led me to the question I’m working on now: how groups should reason about complex systems when the stakes are high and the clock is running. Supply chains are the first domain, but the question applies wherever collective decisions about physical systems need to be made quickly, grounded in evidence, and with reasoning that others can inspect and challenge. The tools I’m building are a bet that this question has a technical answer, and that the answer is buildable.

Thurston model by Adam majewski, CC BY-SA 3.0

Elsewhere

I’m on LinkedIn, where the profile picture is guaranteed to be outdated. My academic publications are on Google Scholar.

Contact

I’m always interested in hearing about new opportunities, collaborations, or a conversation about any of the above.

• Email: elias.jw.public@gmail.com
• LinkedIn: linkedin.com/in/eliasjw
• Location: London, UK