Hang Ten’s bet: AI-native services, not armies of engineers
Hang Ten Systems is an AI-native enterprise services startup founded by former Infosys CEO Vishal Sikka to prove that large-scale enterprise software can be built, changed, and run by AI-first systems instead of the conventional headcount-heavy outsourcing model.
Vishal Sikka has launched Hang Ten Systems with a USD 32 million (approx. RM148 million) seed round led by Mayfield, with a strategic investment from Aramco Ventures and participation from angel investors. This is not a concept-stage AI consultancy; Hang Ten is already working with large customers including Siemens Gamesa Renewable Energy and Fresenius on AI-native project delivery. The startup’s premise is blunt: AI can now perform much of the software customisation, integration, and maintenance work that enterprises have been paying traditional IT services firms billions of dollars for over decades. In other words, Hang Ten’s product is a direct challenge to the $250 billion-plus software services market that has long thrived on labour arbitrage and long projects.
From funding OpenAI to disrupting Infosys’ playbook
This is not Sikka drifting into a trend; it is the continuation of a long-running argument he has been making about AI and enterprise software. He was at the helm of Infosys when it gave a grant to help OpenAI get off the ground, placing him early at the intersection of enterprise IT and frontier AI research. After leaving Infosys in 2017, he founded Vianai Systems, an enterprise AI company seeded with USD 50 million (approx. RM231 million), focused on human-centred AI for large organisations.
Hang Ten is a new venture, distinct from Vianai but with the same fixation: how to make enterprise AI adoption concrete rather than stuck in pilots. Sikka has watched teams who “reach in minutes what could take years of toil” with AI and many more who fail to get value or even cause harm. That experience shows in Hang Ten’s positioning. The company pitches itself as a way for enterprises to make meaningful use of AI as a “force multiplier,” not as a side experiment. The message to incumbent service providers is pointed: the same person who once protected your model is now building to dismantle it.
What “AI-native enterprise services” really change
Hang Ten’s AI-native enterprise services model is structured to attack the core assumptions of legacy IT services. The company describes its approach as agentic code generation, a reusable skills library, and an expert forward-deployed engineering bench that focuses on domains like enterprise transformations, finance, HR, and new product development. In plain terms, recurring software work—customisation, integration, maintenance—is shifted from being mainly human-driven to being orchestrated by AI agents, with specialists supervising and extending them.
The economic logic is the real disruption. Mayfield’s Navin Chaddha says Hang Ten’s AI-native model scales through accumulated project leverage rather than headcount growth. That is the opposite of the traditional IT services playbook, where revenue expansion typically tracks hiring and utilisation. Hang Ten is betting that AI changes unit economics faster than incumbents can adapt: fewer engineers, shorter delivery cycles, and more reusable AI skills per project. If that works, the usual justification for massive offshore teams and layered project management looks less like efficiency and more like legacy bloat.
Why now: public-market anxiety meets enterprise AI ambition
Hang Ten’s timing is opportunistic and, arguably, provocative. Indian software services stocks are under pressure as investors try to estimate how much revenue is exposed to AI tools that can automate coding, testing, documentation, and support. One analyst firm has warned that a severe AI disruption case could drive another 30% to 65% valuation derating for parts of the sector. Meanwhile, industry bodies expect the broader technology industry to reach about USD 315 billion (approx. RM1.45 trillion) in FY26, with the software services slice already a roughly USD 250 billion-plus market under AI pressure.
Incumbent leaders are trying to frame AI as an amplifier rather than a threat. Infosys’ chairman has talked about a USD 300 billion to USD 400 billion (approx. RM1.39 trillion to RM1.86 trillion) AI-first services opportunity by 2030 and has lined up partnerships with AI model providers to reassure clients that the company is adapting. That is the right defensive move: they hold deep client relationships, compliance muscle, and trust inside regulated enterprises. But Hang Ten’s existence signals that those same clients now have a credible alternative narrative: instead of paying for AI to modernise a labour-heavy model, they can adopt a service built to be AI-native from day one.
The future of enterprise AI adoption: extension or replacement?
The open question is whether AI-native services firms like Hang Ten will mostly extend the existing ecosystem or cannibalise it. Hang Ten enters a market where analysts are actively debating whether AI will disrupt traditional IT services or expand them, as one major IT firm’s shares have fallen over 35% amid reassessment of the sector’s long-term economics. Sikka’s own framing—help enterprises “do what no one could do before” with AI—suggests he is not interested in being a subcontractor to old models.
For enterprise buyers, the practical impact could be significant. Hang Ten aims to help large organisations continuously build, modify, and operate software through AI agents and reusable skills, which could mean faster cycles, fewer handoffs, and more direct control over systems. It is already serving early customers such as Fresenius and Siemens Energy’s renewable arm, indicating that AI-native project delivery is not hypothetical. Whether Hang Ten becomes a category-defining company or not, its launch makes one conclusion hard to avoid: AI is no longer a sidecar to consulting—it is starting to rewrite how enterprise service delivery itself is designed, priced, and scaled.
