Wonderful's Sako Arts on scaling enterprise AI: why the last-mile delivery counts
Sako Arts brings experience spanning deep learning research, AI startups, consulting, and enterprise implementation. He is currently the CTO of Wonderful NL, a Dutch-Israeli startup focused on enterprise AI that recently made headlines after reaching unicorn status. Sako shared what’s behind their exponential growth and effective strategies for scaling AI on an enterprise level.
Founded in early 2025, Wonderful reached a $5B valuation in under 6 months after raising $550 million in a Series C round. Wonderful initially saw traction as a customer service AI agent platform, but more recently it has pivoted from customer service to applying AI at an enterprise level. The startup combines a multi-model AI platform, local deployment teams, and expert advisors to help enterprises transform workflows and become AI-native.
Wonderful's differentiator
Wonderful focuses on going from isolated AI use cases to organization-wide transformation by building a platform that is compatible with any AI model and existing tech stack.
Sako explains that while their platform is a “technological marvel,” code is very affordable these days and can be replicated easily.
"The real differentiator is our Forward Deployed Model, the fact that we bring our own exceptional people at no cost to do the last mile delivery,” he says. “AI can do it all, and it has been able to do so for months. The gap is in the implementation, which is what we solve and do so quickly.”
Behind Wonderful’s success
Wonderful works across 30+ markets and is growing, including European markets, South American markets, and the Middle East. Sako attributes Wonderful’s rapid global expansion to three factors: product quality, early-on traction, and sufficient capital after multiple successful funding rounds. Beyond that, he adds, exceptional talent allows companies to move at “breakneck speed and maintain momentum”
"You can build a good company with good people, but for an excellent company you need excellent people."
Sako also acknowledges that rapid success comes down to being at the right place at the right time, "so in other words, luck!"
Wonderful’s human success factor is also a challenge going forward for Sako, finding exceptional people to grow Wonderful as quickly as they are onboarding clients.
Advice for running AI in enterprise
Enterprise AI has shifted from individual productivity to embedding AI directly into applications. This often involves agents working across systems, using intelligent orchestration and governed data.
When asked what every enterprise applying AI should ask itself, Sako says companies should ask three questions:
Are their source systems (ERP, CRM, etc.) and their commercial contracts ready for the agentic era?
Will they be able to consume their own data with high frequency and low latency without making every agentic business case break on API costs?
Could they use the power of agentic coding to replace these systems to avoid such issues?
These are worth asking before purchasing any AI service, because they often come down to the same thing: whether the systems underneath are ready for it.
Once those foundations are in place on an executional level, Sako advises teams to stop strategizing and start focusing on key basics to move toward implementation.
“Start implementing, choose the right platform that is actually built for deploying agents, find a really impactful, high-frequency use case where token waste is an afterthought, put the right people with serious engineering skills on the job, involve the business and ensure the engineers understand the why, go to production as quickly as possible and be prepared to iterate fast."
Sako will host a workshop, Building Trusted Enterprise AI Agents, at TechDays on 15 October. Starting from a real enterprise use case, participants follow the full journey from an initial idea to a fully operational AI agent. Register here.
