The man who once ran one of the world's largest IT-services firms now wants to undercut the entire model. Vishal Sikka, chief executive of Infosys from 2014 to 2017, has launched a new startup, Hang Ten Systems, with a $32 million seed round and a thesis that artificial intelligence can do much of the work that the outsourcing industry sells by the hour, according to TechCrunch.

What the company does

Palo Alto-based Hang Ten Systems describes itself as an enterprise AI services company built around "agentic code generation, reusable AI skills, and domain expertise," the company said. In plainer terms: rather than staffing a client project with hundreds of developers and analysts, it deploys AI agents to build, modify and run software, with a smaller team of human specialists supervising.

The startup said it has already signed early customers, including the wind-turbine maker Siemens Gamesa and the medical-services group Fresenius. The $32 million round was led by the venture firm Mayfield, with a strategic investment from Aramco Ventures, the investment arm of Saudi Aramco, Inc42 reported. Yahoo co-founder Jerry Yang sits on the board. The company said it had been operating for only about a month when it announced the raise.

What Sikka is challenging

To see the bet, it helps to understand how the incumbent industry works. Firms such as Infosys, Tata Consultancy Services, Wipro, HCL and Accenture win multi-year contracts to run clients' technology: maintaining old systems, writing new software, managing data centers and staffing help desks. They bill either by time and materials — so many engineers at so many dollars an hour — or at a fixed price for a defined scope. Either way, profit is largely a function of how many people are deployed and the margin between what they cost and what the client pays.

That model built one of the largest employment engines in the modern economy. India's technology and business-process sector employs roughly 5.8 million people, according to the industry body NASSCOM, and the global market is measured in hundreds of billions of dollars a year. Infosys alone reported about $18.6 billion in revenue in its 2024 fiscal year.

The model's strength — it scales by adding people — is also its exposure. If AI agents can write, test and deploy software at a fraction of the human cost, the case for a 500-person delivery team weakens.

Disruption, or a bigger market?

The industry is not conceding the point. Infosys's own leadership has argued that AI will expand the addressable market rather than shrink it, pushing it toward $300 billion to $400 billion by 2030 as cheaper automation makes more digitization economical — much as cloud computing ended up enlarging, not destroying, outsourcing revenue. Analysts are more divided; some, including at Jefferies, have flagged "meaningful AI disruption" risk to the services model, TechCrunch noted.

Sikka is firmly in the disruption camp. Hang Ten is not pitched as an add-on to the existing model but as an alternative to it.

Why it matters

Sikka brings unusual credibility to the argument. He led Infosys, and before that spent more than a decade at SAP rising to chief technology officer — meaning he helped build the enterprise software the services firms are paid to implement. His earlier startup, the enterprise-AI company Vianai, raised substantial funding including from SoftBank's Vision Fund 2.

The $32 million is small next to firms worth tens of billions, but it is enough to suggest experienced backers see an opening — and Aramco's involvement is notable, since large energy companies are among the biggest buyers of IT services, implying at least one major potential customer is hedging. Whether Hang Ten can scale fast enough to dent the incumbents is unproven. But its launch sharpens an uncomfortable question now circulating inside every large services firm: if AI can do the work, what exactly are clients paying for?