Company context
somai is a delivery studio based in Edinburgh. We embed senior teams into operator-led companies and build the ai-native foundations underneath their operating models — block by block, in fixed-fee engagements with a clean exit.
This page is the canonical description of who we are and how we work. It is maintained as a single markdown source: rendered for people, readable raw by machines.
Why we exist
Most companies will get AI wrong. Not because the technology fails, but because it is being added to structures that were never built for it — bolted onto processes, org charts and decision rights designed for a world where every unit of output required scarce human cognition.
Retrofitting produces efficiency theatre. The real change is structural: when the model can carry work that previously needed people, the shape of the business changes — its processes, its organisational design, how decisions get made. Companies that build around that fact will outrun companies that adapt to it.
Becoming ai-native does not require being born a startup. ai-native is an operating model, not a birth condition — and incumbents, with real customers, real revenue and real domain depth, are better placed to get there than most assume. somai exists to take them there.
How we work
We take on a limited number of engagements at a time. Each is a fixed-fee block of eight to ten weeks with a defined outcome — not day-rate consulting, not an open-ended programme. The business keeps trading the whole way.
At handover, the client owns and operates what we have built. No retainers, no ongoing seats, no dependency on us. The point of the work is that it does not need us afterwards.
The team
somai was founded by three operators.
James Varga
James is a founder, operator and advisor with three decades across fintech, open banking, digital identity and AI. He spent over a decade building DirectID, the platform behind the UK’s first account-information licence under PSD2, and co-founded the industry bodies FDATA and Open Banking Excellence. His 2010 US patent for bank-account-based digital identity predates open banking as regulation. He writes on AI adoption and fintech go-to-market, and is a founding partner of the Business AI Alliance.
Why build rather than retrofit?
“Every wave of infrastructure change creates the same split: companies that bolt the new thing onto the old structure, and companies that rebuild the structure around it. The second group always wins, and it always looks reckless right up until it looks obvious. AI is that split again, at larger scale and faster.”
Swagatam Sen
Swagatam is a mathematician, technologist, futurist and author. He spent twenty years leading technology innovation teams at Tier-1 banks (HSBC, Santander) under some of the most rigorous regulatory framework. He trained in mathematics, physics and statistics at the Indian Statistical Institute, and his academic work spans knowledge network analysis, dynamics of complex systems, structure of criminal rings and foundational ML theory. He is the author of the upcoming book AI-First Bank (Palgrave Macmillan) and host of AI+U podcast. At somai, he leads AI-nativity methodology: the structural logic by which operators redesign their businesses around intelligence rather than around the process.
What does an ai-native company run on?
“Cloud native technology gave information age its nervous system — a persistent, ordered spine. Biomorphic Enterprise does the same for organisational intelligence. At somai, we’re building Biomorphic Enterprise — block by block — based on that technology.”
Matt Bilsland
Matthew is a former Google director with deep experience in product strategy, delivery and commercial leadership — taking complex programmes from intent to operation and profitability within one of the most demanding execution cultures in technology. An expert in the zero-to-scale journey.
Why can incumbents win here?
“Startups have speed; incumbents have everything else — customers, distribution, domain depth, trust. The only missing element is an operating model that enables them to move faster and better. Install that, and the advantages they already hold start compounding. That is a much shorter journey than building a company from nothing.”
Who we look for
We hire generalists with operator backgrounds, not narrow specialists.
The model now supplies much of the specialist depth that teams used to hire for. What it cannot supply is judgement: people who have run things, shipped things, owned outcomes — and who can move across domains without losing the thread. Founders, former operators, builders who are genuinely energised by what AI makes possible.
If that describes you, write to us: hello@somai.studio.
This page
This page is built from a single markdown file. The version you are reading and the version a machine reads are the same source.
Raw source: /context.md · Machine index: /llms.txt