Dr. Giulia Marcucci spent years working at the intersection of research and applied technology, developing a new way to run artificial intelligence using light. The idea was simple in principle but complex in execution. If AI could be processed through photonics instead of traditional computation, it could drastically reduce energy use while increasing speed and efficiency.
That idea became LumiAIres, a deeptech company building neuromorphic photonic processors designed to solve one of AI’s biggest challenges: the cost of computation.
The journey from research to company was not straightforward. It required stepping away from academia, building a team from scratch, and navigating a very different world of funding, markets, and commercialisation.
Moving Beyond Research
The core technology behind LumiAIres was developed over nearly a decade of research. But at a certain point, Dr. Giulia realised the work was no longer just academic. That there was a clear path to building a product.
Leaving the University of Glasgow was a turning point. Without forming a traditional spin-out, the team had to build independently, without many of the structures that typically support deeptech founders.
This created an early set of obstacles. Many programmes were designed for either early researchers or software startups. Hardware companies, especially those building at the edge of physics and computing, often sat in between.
However, she was not making the journey alone. Her co-founder, Jeremy Delhommeau, with experience in manufacturing and programming, had been part of the journey from the beginning. Where Giulia brought deep technical insight, Jeremy focused on translating that into something tangible. Together, they began turning an idea into a product that could exist outside of a research environment.
Understanding The Reality Of Deeptech
Very quickly, the realities of building a hardware company became clear.
Unlike software, progress is slower and often invisible in the early stages. Building infrastructure, validating technology, and preparing for real-world use all take time. Costs are also significantly higher from the outset.
“Building a lab alone costs over 1 million, and a product will not be ready for three years” - Dr. Giulia Marcucci
This shaped how the company approached its early stages. Funding came in small steps, starting with initial grants and support, followed by further backing as the technology and its potential became clearer.
At the same time, the team had to stay close to the market. Their first instinct was to target hyperscalers and large technology companies. Over time, that changed. Through conversations and validation, they identified stronger early opportunities in space and defence. That shift gave the company a clearer path forward and opened up new funding routes.
Finding The Right Environment
The early stages of building the company were also shaped by access to the right people and perspectives.
Giulia first came across CodeBase through a local event. That led to joining Techscaler Catalyst, which introduced structured ways of thinking about customers, markets, and positioning.
More importantly, it created a sense of being part of something wider. Coming from academia, that mattered. Building a company can be isolating, especially in deeptech, where few people are working on similar problems.
The Techscaler Silicon Valley programme became a turning point.
Being in that environment changed how the company approached conversations, from how they spoke to investors to how they thought about growth. It also led to tangible outcomes, including new investor relationships and insightful first conversations with potential customers.
"You meet cool people, you meet sharks, you meet somebody in the between. But at the end of the day, you are exposed to everyone when you go there (Silicon Valley)" - Dr. Giulia Marcucci
Building Momentum
As these pieces came together, progress began to accelerate.
The company moved closer to a commercial prototype, with a focus on validating the technology in real-world environments. At the same time, the team began to grow.
They also progressed from early funding into a larger seed round, with growing investor confidence supported by clearer positioning and stronger validation.
This stage of the journey looks very different from where they started. What began as research is now moving towards deployment.
Founder Takeaways
Giulia’s experience reflects a set of realities that many deeptech founders will recognise.
• The transition from research to company is as much about mindset as it is about technology
• Hardware requires patience, both in development and in funding
• Early market assumptions will change, and that is part of the process
• Being around other founders helps make difficult stages more manageable
• Exposure to global ecosystems can reshape how a company thinks about growth
What’s Next
LumiAIres is now focused on validating its technology and preparing for wider deployment.
The direction is clear: build systems that allow AI to run faster, with less energy, and in environments where traditional infrastructure cannot operate.
The journey from research to company is still ongoing. But the foundation is now in place, and the next phase is about scale.
.png)
