In an era where the cost of developing AI is skyrocketing, London-based startup Inephany is carving a disruptive path with its intelligent optimisation platform for neural networks. With a fresh £1.8 million pre-seed round led by Amadeus Capital Partners, and backed by Sure Valley Ventures and AI pioneer Professor Steve Young, the company is poised to transform how Large Language Models (LLMs) and other neural architectures are trained and deployed.
The investment will be channelled into three strategic priorities: growing the core engineering and research team with top-tier talent, advancing the AI optimisation engine to handle more complex and diverse models, and supporting the company’s first wave of enterprise adopters.
Inephany’s answer to the AI compute crisis
The exponential growth of AI has ushered in an era of unprecedented computational demands. Since 2012, AI computational power has been doubling approximately every 3.4 months, far outpacing Moore’s Law, which predicted a doubling every two years.
Training advanced models like GPT-4 is estimated to have cost between $60 million and $100 million, and projections for next-generation models suggest expenses could reach up to $1 billion. This rapid increase in computational demand highlights the unsustainable nature of traditional training and optimisation methods.
Inephany addresses this challenge with its innovative AI-driven optimisation system, which intelligently manages the training process in real time. It paves the way for scalable and sustainable AI development, offering at least ten times more cost-effective solution.
The team behind this innovation
Inephany was founded in 2024 by AI veterans, including Dr. John Torr, a former machine learning engineer at Apple Siri; Hami Bahraynian; and Maurice von Sturm, co-founders of conversational AI startup Wluper. The team has deep AI expertise, spanning speech recognition, dialogue systems, and neural network training. With a bold mission to rethink how models learn and evolve, Inephany aims to enable AI’s more efficient, scalable, and environmentally conscious future.
The founders were collectively frustrated with the inefficiencies of current AI training methods. Having seen firsthand how compute-heavy and unsustainable model development had become, they envisioned a platform that could make training smarter rather than just more powerful, optimising performance through intelligent guidance rather than brute-force scaling.
A revolutionary approach to optimisation
Unlike conventional approaches that rely on extensive trial-and-error, Inephany’s technology enhances sample efficiency, accelerates training timelines, reduces development durations, and improves overall model performance, while significantly cutting compute expenses.
This system enables neural networks to learn more quickly and with fewer computational resources, unlocking at least 10x cost savings. It reduces not only the energy footprint of training models like GPT-4, but also the time and resources needed to iterate and deploy.
Although Inephany initially focuses on training-time optimisation for LLMs, its technology is designed to be model-agnostic. The platform has promising applications in other architectures, such as Recurrent Neural Networks (RNNs) for financial forecasting, and Convolutional Neural Networks (CNNs) for computer vision in autonomous vehicles. The company also plans to extend its optimisation capabilities to inference-time compute, enabling end-to-end efficiency gains across the AI lifecycle.
Our thoughts
Inephany is a deep tech startup with technical depth, commercial relevance, and a sustainable mission. As compute costs soar and AI development becomes more exclusive, its approach democratises access and improves global innovation capacity. With top-tier investors, a stellar founding team, and broad applicability, it’s positioned to become a foundational player in the next wave of AI evolution.
John Torr, CEO at Inephany, said: “We are thrilled to be backed by such experienced investors, and having a seasoned entrepreneur and AI pioneer like Professor Steve Young as our chair is a true privilege. Current approaches to training LLMs and other neural networks are extremely wasteful across multiple dimensions. Our unique solution tackles this inefficiency head-on, with the potential to radically reduce both the cost and time required to train and optimise state-of-the-art models. As we prepare to deliver our first products later this year, we are incredibly excited to embark on the next chapter of our journey—and to help shape the ongoing AI revolution by transforming AI optimisation.”
Amelia Armour, Partner at Amadeus Capital Partners, said: “We very much look forward to backing John, Hami, and Maurice as they tackle key efficiency challenges in current AI training. Their innovative approach to automating and optimising neural network training has the potential to reduce costs by an order of magnitude and accelerate advancements across AI applications. If rolled out at scale, the impact of this on what models can deliver will be very substantial.”
Professor Steve Young said: “As the use of AI spreads ever wider, moving beyond the traditional applications of speech, language and vision into new and diverse areas such as weather prediction, healthcare, drug discovery and materials design, the need for very efficient training of accurate neural models is becoming critical. The groundbreaking new approach being developed by Inephany marks a step change in neural model training technology and I am delighted to join the team as chair and investor.
