A company founded by four University of Edinburgh graduates has raised $5.3m in seed funding for its technology that helps organisations accurately predict how much compute they need, potentially saving millions in both costs and carbon emissions.
AI infrastructure startup Expanse, founded in October 2025 while two of the four founders were still students, has built software that accurately predicts what an AI workload requires before it runs.
Currently, engineers must estimate how many GPUs, CPUs, how much memory and how long their code will run before execution begins. Those estimates determine how billions of dollars of infrastructure are allocated every day. Industry estimates suggest approximately 30% of cloud spend is lost through over-allocation, while Microsoft Research has reported GPU utilisation of around 50% across many internal deep learning workloads.
Expanse recovers wasted GPU capacity in AI computing clusters by predicting job resource needs, suggesting efficiency tweaks, and catching failing jobs early — helping organisations get more AI work done from hardware they already have, instead of building new data centres.
“Today, every AI workload begins with an educated guess,” said Ismaeel Bashir, co-founder and CEO of Expanse. “Engineers shouldn’t have to predict exactly how much compute their code will need before they press run. The machine should carry the uncertainty, not the person.”
The timing reflects a growing challenge across the industry. The scale of computing required to run AI programmes necessitates ever larger data centres for storing and processing information, causing environmental concerns over the energy required, including for cooling. Optimising compute efficiency will help data centres run more sustainably.
The seed funding round was led by Crane Venture Partners, with participation from PXN Ventures and angel investors including former DeepMind researchers and leaders from AI infrastructure teams.
Ismaeel Bashir developed a multimodal HPC resource prediction system whilst a masters student at the University’s EPCC (formerly known as Edinburgh Parallel Computing Centre). Together with co-founders – his friends and fellow students Chief Technical Officer Nikodem Bieniek, Chief Product Officer Eren Mendi and Chief Operating Officer Yafet Melake – he recognised that even the world’s most sophisticated compute environments still relied on manual estimation to allocate resources efficiently.
Expanse received company formation support from Edinburgh Innovations (EI), the University’s commercialisation service, including its first finance, as winners of EI’s Launch Point competition in September 2025. Earlier this year, they were accepted on to prestigious startup accelerator Y Combinator, in San Francisco, where the team is now based.
Mr Bashir said:
When you start a company everyone thinks you’re delusional, and you’re spending money out of your own pocket. Being advised by EI and winning the competition, even though it was a small amount of money, was our first validation, and we will always be extremely appreciative of the University’s support. ”
Dr Andrea Taylor, CEO of Edinburgh Innovations, said:
Our venture creation teams, alongside Old College Capital, our venture investment fund, support the most exciting ideas and technologies emerging from the University.
It’s fantastic to see our innovators addressing some of the world’s most pressing issues, such as how we will harness data and AI for good whilst tackling the climate and environmental crisis.
We wish the Expanse team every success.”
