5 Key Takeaways
- Vishal Sikka's startup Hang Ten Systems uses AI-driven agentic code generation to replace human IT services labor, breaking the traditional linear-scaling model.
- The company has raised a $32 million seed round and already has early customers like Siemens Gamesa Renewable Energy and Fresenius.
- Sikka's deep industry experience as former Infosys CEO and SAP CTO gives his venture credibility and insider knowledge of the IT services sector.
- Hang Ten challenges the $400 billion IT services industry, sparking debate about whether AI will disrupt or expand the market.
- The startup's approach leverages reusable AI skills and domain expertise, allowing its knowledge base to grow with each project without hiring more people.
Can AI Replace the World's IT Services Armies? Vishal Sikka Puts $32 Million on the Line to Find Out
The former Infosys CEO launches Hang Ten Systems, an AI-native services startup that aims to do for enterprise software delivery what generative AI did for content creation — automate it at scale.
The global IT services industry is an invisible colossus. For decades, it has quietly powered the digital operations of corporations around the planet by taking on the unglamorous but indispensable work of customizing, integrating, and maintaining enterprise software. Now, one of the most influential figures ever to lead that industry is betting that artificial intelligence can make much of that human labor obsolete. Vishal Sikka, the former CEO of Indian IT services giant Infosys, has launched a new startup called Hang Ten Systems, armed with a $32 million seed round and a conviction that AI-driven code generation will reshape how companies build and run their software.
Hang Ten Systems emerged from stealth on Wednesday with a mission to help enterprises "continuously build, modify, and operate software using AI-driven development and automation." It is a direct challenge to the business model that made Infosys and its rivals hundreds of billions of dollars. Sikka, who led Infosys from 2014 to 2017, is not merely dipping a toe into the AI wave; he is trying to ride it to the heart of an industry that has defined his career. In his own blog post announcing the venture, the 59-year-old executive wrote that Hang Ten is already assisting large enterprises to "hang ten on the biggest wave of our lifetimes."
The announcement comes at a moment of profound uncertainty for the IT services sector. Companies that once thrived on a simple formula — hire legions of engineers, assign them to client projects, and bill by the hour or by the head — are now grappling with generative AI models that can write code, generate documentation, and troubleshoot systems autonomously. Some analysts argue the industry could be among the first to face genuine upheaval from AI. Others, including Infosys's current chairman Nandan Nilekani, maintain that the technology will open up vast new markets. Sikka's new venture puts those competing narratives to the test.
From Outsourcing Machine to AI Native
To understand what Hang Ten Systems attempts, it is helpful to grasp what IT services firms actually do. Large organizations rely on complex software platforms — such as those made by SAP, Oracle, or Salesforce — to run everything from accounting to supply chains. These platforms rarely work "out of the box" for a specific business. They need to be tailored, integrated with other systems, upgraded over time, and constantly maintained. IT services companies deploy teams of consultants and developers to perform that work on the client's behalf.
For decades, the model scaled in lockstep with headcount. If a services firm wanted to take on more projects or win larger contracts, it would hire more people. This linear relationship between revenue and employees made the industry a massive employer, particularly in India, and generated enormous fortunes. Sikka understands that engine intimately. Before his stormy tenure at Infosys, he spent 12 years at SAP, rising to chief technology officer and becoming the architect of the HANA in-memory database, a product that redefined the company's core technology. He later served on the board of Oracle. Few people on the planet possess his depth of experience in both building enterprise software and delivering it as a service.
Hang Ten is designed to break the linear-scaling constraint. The startup describes itself as an enterprise AI services company built around "agentic code generation, reusable AI skills, and domain expertise." In simpler terms, instead of dispatching a roomful of developers to write custom code, Hang Ten deploys artificial intelligence agents that can interpret a client's requirements and generate or modify software automatically. These agents draw upon a library of pre-built components that capture expertise from specific industries, so the system gets smarter and more efficient with every project.
"Traditional services scale linearly with headcount. Hang Ten is built so its leverage grows with every project." — Mayfield, the venture capital firm that led the $32 million seed investment
Mayfield managing partner Navin Chaddha said the company "just got started a month back" and already has customers. That early traction is notable because it suggests that at least some enterprises are willing to entrust their vital software pipelines to a startup that has replaced much of the conventional human workforce with AI. The first publicly named customers are Siemens Gamesa Renewable Energy and Fresenius, two large industrial and healthcare companies that would otherwise be prime targets for traditional IT services firms.
A Team Forged Over Decades
Hang Ten is headquartered in the San Francisco Bay Area and is actively hiring across delivery, engineering, sales, and leadership. The early crew, according to their LinkedIn profiles, is a tight-knit group of executives who have worked with Sikka for years at SAP, Infosys, and his previous enterprise AI startup, VianAI. Co-founder Navin Budhiraja serves as chief technology officer; Sanjay Rajagopalan is chief design officer; and Tao Liu is senior vice president of forward deployed engineering. The presence of such a seasoned team signals that Sikka has drawn deeply from his professional network to build a company capable of executing on an ambitious technical vision.
The startup's board includes Yahoo co-founder Jerry Yang, a name that lends additional credibility to the venture's web-scale aspirations. Strategic funding also came from Aramco Ventures, the investment arm of Saudi Aramco, a company with sprawling, mission-critical IT needs that epitomize the kind of client Hang Ten aims to serve.
Hang Ten's debut is Sikka's second act as an AI entrepreneur. After stepping down as Infosys chief executive in 2017 — a departure that followed a public dispute with the company's founders — he founded VianAI. That startup emerged from stealth in 2019 with a $50 million seed round and later raised $140 million in a 2021 funding round led by SoftBank Vision Fund 2. VianAI focused on enterprise AI applications and analytics, building tools that helped businesses use artificial intelligence to make better decisions. Chaddha was careful to distinguish the two ventures, explaining to reporters that Hang Ten is aimed at a completely different market.
While VianAI sold software, Hang Ten sells outcomes. It is positioned as an AI-native services company — one that doesn't just hand over a product but takes responsibility for delivering and operating the entire solution. This distinction is critical. Enterprises that buy IT services are not just purchasing code; they are buying peace of mind, assured that their systems will be up, secure, and compliant. Sikka's wager is that AI agents can provide that same assurance with far greater efficiency than a human-staffed team ever could.
A $400 Billion Question Mark
Hang Ten's emergence sharpens an already heated debate on Wall Street and in corporate boardrooms about the future of IT services. Analysts at investment bank Jefferies argued earlier this year that the sector may be among the first to face meaningful disruption from artificial intelligence. Large language models have demonstrated a startling ability to produce software code, summarize technical documents, and even diagnose system faults. If those capabilities continue to improve, the reasoning goes, the core value proposition of the services industry — smart people solving technical problems at scale — could erode.
Infosys, not surprisingly, sees a glass half full. Its chairman, Nandan Nilekani, said publicly this week that AI is likely to expand the industry's addressable market. Rather than cannibalizing existing services, the technology will enable entirely new classes of work that were previously too complex or too expensive to automate. Infosys has been telling investors that "AI-first services" could represent a $300 billion to $400 billion market by 2030. The company is busily forging partnerships with frontier AI labs such as Anthropic and OpenAI to ensure it remains relevant.
Yet the market is signaling deep unease. Infosys shares have fallen more than 35% this year, a decline that reflects investor anxiety about whether existing business models can withstand the AI onslaught. If a startup like Hang Ten can deliver the same — or better — software services with a fraction of the headcount, traditional providers may face severe margin compression or outright obsolescence. Sikka's intimate knowledge of those providers' inner workings makes his startup a particularly formidable challenger. He understands not only the technological levers but also the sales cycles, the client relationships, and the cultural inertia that have protected the incumbents for so long.
What Makes Hang Ten Different
The term "agentic code generation" might sound like jargon, but it points to a genuinely novel approach. Traditional automation in IT services has relied on scripts and rigid rules that require extensive human setup and maintenance. Agentic code generation, by contrast, uses AI systems that can pursue goals autonomously: understand a business requirement, break it down into technical tasks, write and test code, and then integrate the result into a running system. Such agents can also monitor and modify the software over time as requirements evolve or bugs appear. If you have ever interacted with a chatbot that can generate functional software snippets, imagine that ability scaled up, guided by deep industry templates, and hardened for the security and compliance demands of a multinational corporation.
The "reusable AI skills" Hang Ten mentions are another key differentiator. Every time the system completes a project for, say, a renewable-energy firm, it captures patterns, data models, and integration know-how that can be repurposed for the next client in that sector. Over time, the startup's knowledge base grows without hiring more people. This is the leverage Mayfield highlighted. It is a self-reinforcing cycle: more projects lead to a richer skill library, which leads to faster and cheaper delivery, which attracts more projects.
🔑 The Self-Reinforcing Cycle
More client projects → Richer AI skill library → Faster, cheaper delivery → More competitive pricing → More client projects. All without scaling headcount linearly.
Of course, the technology must prove itself at scale and under real-world pressure. Enterprise software environments are messy, filled with legacy code, homegrown integrations, and security policies that can trip up even seasoned engineers. AI agents have made impressive strides in controlled settings, but whether they can handle the chaos of a live enterprise is an open question. Sikka's decision to name Siemens Gamesa and Fresenius as early clients suggests he is confident enough to let the market watch.
What Happens Next
Hang Ten is still in its earliest days. It has seed funding, a compelling founding team, and a handful of initial customers. The path forward involves hiring the right mix of AI researchers, engineers, and client-facing experts, while maintaining the culture of rapid iteration that distinguishes a startup from the slow-moving giants it hopes to disrupt. The company plans to expand across multiple global locations to meet enterprise demand, presumably establishing delivery centers that blend AI-driven automation with just enough human oversight to guarantee quality and handle exceptions.
Sikka's move also sends a broader signal to the technology industry. After years of speculation about whether AI would augment or replace knowledge workers, one of the world's foremost experts on enterprise software is placing a sizable bet on replacement. He is not launching a tool that helps existing IT services companies do their jobs better; he is launching a competitor that aspires to do the whole job differently. That entrepreneurial leap, backed by sophisticated investors, may force enterprise customers and IT services providers alike to accelerate their own AI strategies.
For the millions of engineers employed in the IT services sector, Hang Ten represents both a threat and a promise. The threat is obvious: if AI can indeed handle large swaths of custom development and maintenance, the demand for human coders in those roles could shrink. The promise, less visible but no less real, is that the industry may evolve into higher-value activities — architecting complex systems, solving novel business problems, and guiding AI agents rather than writing every line of code. Sikka's previous venture, VianAI, was ostensibly about harnessing AI to improve decision-making. Hang Ten goes further by embedding AI directly into the production line of enterprise software delivery.
The giant question hanging over the startup is the same one that hangs over the entire industry: can AI deliver the consistency, reliability, and trust that enterprises require? A billing system failure during a holiday shopping weekend or an outage in a hospital's patient-records platform can cost millions and endanger lives. IT services companies have spent decades building safeguards, escalation paths, and accountability frameworks. Hang Ten will need to match or exceed that reliability while operating on a fundamentally different economic model. Its ability to do so will determine whether Sikka rides the wave or gets swept away by it.
In the meantime, the company's emergence has already reshuffled the conversation. Silicon Valley investors, who have poured billions into foundation model companies and AI infrastructure, are now looking for applications that can upend established industries. Mayfield's backing of Hang Ten signals that a respected, sixty-year-old venture firm sees IT services as one of the ripest targets. And Aramco Ventures' strategic investment hints that even the largest enterprises are keen to explore alternatives to the traditional vendor model.
Sikka, for his part, seems to relish the collision of his past and his future. He once stood at the apex of the very system he now aims to supplant. He knows its weaknesses better than almost anyone. With $32 million in funding, a battle-tested inner circle, and a vision of AI-native services, he is stepping back into the arena. The IT services industry has rarely seen a challenger who so thoroughly understands the terrain. Whether it will yield or fight back remains to be seen, but one thing is clear: the debate over AI's role in enterprise software is no longer theoretical. It now has a name, a face, and a growing list of customers. The wave is here, and Vishal Sikka is hanging ten.
No comments:
Post a Comment