Vishal Sikka is back in the arena.
After a long career that helped define how large enterprises think about software, services, and—more recently—how they operationalize AI, the former Infosys chief has launched a new venture with a clear ambition: challenge the IT services world at a time when the category itself is being rewritten. The company is backed by Mayfield and Aramco Ventures, and it brings together veterans from SAP, Infosys, and VianAI—an unusual mix that signals the startup isn’t simply trying to “add AI” to existing consulting playbooks. Instead, it appears designed to attack the underlying economics of services delivery: speed, cost, and the gap between what enterprises buy and what they actually deploy.
The timing matters. Traditional IT services have been under pressure for years—first from cloud migration, then from automation, and now from generative AI. But the pressure isn’t uniform. Some parts of the services market are being commoditized, while others are becoming more valuable because they sit closer to business outcomes: process transformation, data modernization, and systems integration that turns enterprise software into something that can adapt. In that context, a new entrant led by someone with deep experience in both enterprise software and large-scale delivery could find a narrow but high-impact opening.
What makes this launch particularly interesting is the composition of the team and the implied strategy behind it. Veterans from SAP bring an intimate understanding of enterprise application landscapes—ERP, finance operations, supply chain processes, and the realities of implementing and extending them in complex organizations. Infosys alumni bring experience with large-scale delivery models, governance, and the mechanics of running service operations across geographies and industries. And VianAI’s presence suggests a stronger emphasis on AI-native approaches—likely including how AI systems are built, evaluated, and integrated into enterprise workflows rather than treated as standalone demos.
That combination points to a thesis that many incumbents struggle to execute cleanly: AI won’t replace enterprise software or services overnight, but it will change how those services are delivered. The question is whether a startup can build a delivery engine that is meaningfully different—one that reduces friction for customers and improves reliability for the vendor—while still meeting the compliance, security, and operational standards that enterprise buyers demand.
A challenge to the IT services “default”
For decades, IT services have operated on a fairly recognizable pattern. Enterprises identify needs, vendors propose roadmaps, teams implement systems, and ongoing support keeps things running. Even when automation is introduced, the structure often remains the same: work is broken into deliverables, staffed by specialists, and managed through project cycles.
Generative AI disrupts that model in two ways. First, it accelerates certain tasks that used to require manual effort—documentation, code scaffolding, test generation, and knowledge retrieval. Second, it changes the nature of “requirements.” When AI can draft, summarize, and propose options, customers may expect faster iteration and more interactive discovery. That shifts value away from long planning cycles and toward continuous refinement.
But there’s a catch: AI also introduces new risks. Enterprises don’t just need outputs; they need traceability, auditability, and predictable behavior. They need systems that can be monitored, governed, and improved over time. That means the real opportunity isn’t simply using AI to do work faster—it’s building a services delivery system that can incorporate AI safely and consistently.
This is where a startup can differentiate. If the venture is built around an AI-enabled delivery pipeline—one that standardizes how knowledge is captured, how solutions are validated, and how changes are tested—then it can offer customers something that traditional services often struggle with: shorter time-to-value without sacrificing quality.
The backing also hints at seriousness. Mayfield is known for supporting companies that build durable technology platforms rather than only services businesses. Aramco Ventures brings a strategic lens rooted in industrial scale and operational complexity. That combination suggests the startup may be aiming beyond “consulting with AI” and toward a repeatable productized approach—something that can be deployed across customer environments with less bespoke effort.
Why enterprise software veterans matter
It’s easy to underestimate how much enterprise software implementation depends on institutional knowledge. SAP environments, for example, aren’t just databases and modules—they’re living systems shaped by years of configuration, custom extensions, integrations, and governance. Many organizations have multiple layers of complexity: legacy processes, regulatory constraints, and business units that interpret “standard” differently.
When a startup assembles veterans from SAP and Infosys, it’s not just about credibility. It’s about reducing the learning curve that typically slows down new entrants. Enterprise buyers are wary of vendors who can build prototypes but can’t operate at production scale. They want teams that understand how to handle edge cases, how to manage change control, and how to ensure that performance and security requirements are met.
The presence of VianAI veterans adds another dimension: AI isn’t just a feature; it’s a capability that must be engineered. In enterprise contexts, AI systems must be integrated into existing workflows—ticketing, incident management, knowledge bases, development pipelines, and operational dashboards. They must also be evaluated against metrics that matter to businesses: accuracy, latency, cost per task, and failure modes.
If the startup’s team truly spans these domains, it can potentially build a delivery model that treats AI as infrastructure rather than an add-on. That would be a meaningful shift from many current offerings, where AI is often layered onto existing processes without changing the underlying operating model.
The “next phase” of IT services competition
The IT services market is often described as mature, but it’s not static. It’s evolving in at least three directions:
1) Automation of delivery tasks
Many vendors already use automation tools, but generative AI raises the ceiling. The difference between incremental automation and a new delivery engine is whether the vendor can reduce dependency on specialized labor for routine tasks while maintaining quality. That requires strong internal tooling, standardized templates, and robust validation.
2) Outcome-based engagements
Enterprises increasingly want measurable outcomes—reduced cycle times, improved compliance, lower operational costs, better forecasting, fewer incidents. Vendors that can tie their work to these outcomes gain leverage. Startups can sometimes move faster here because they’re not constrained by legacy delivery structures.
3) Integration and data modernization
AI initiatives fail when data is fragmented or systems are disconnected. The most valuable services increasingly involve integration: connecting ERP, CRM, data warehouses, and operational systems into a coherent architecture. This is where enterprise software expertise becomes essential.
A startup led by someone like Vishal Sikka—who has previously navigated the intersection of enterprise software and large-scale delivery—could position the venture to compete in these areas with a more modern approach. The unique take is not just “we’ll use AI,” but “we’ll redesign the delivery workflow so AI becomes a lever for outcomes.”
What customers will likely care about first
Even if the venture has ambitious goals, early customer adoption will depend on practical concerns. Enterprises don’t buy vision; they buy risk reduction and execution confidence. In the first wave of deals, buyers will likely evaluate:
Reliability and governance
How does the system behave when it’s uncertain? What guardrails exist? How are outputs validated? Can the vendor provide audit trails and explainability where needed?
Security and compliance
Enterprise environments require strict controls. The startup’s ability to integrate with existing security frameworks—identity management, access controls, logging, and data handling policies—will be critical.
Integration depth
Customers will ask: can you plug into our SAP landscape, our CI/CD pipelines, our data stores, and our operational tooling? Or do we need to re-architect everything?
Cost predictability
AI can be expensive if not managed. Buyers will want clarity on cost per workflow, compute usage, and how the vendor prevents runaway expenses.
Time-to-value
The best AI-enabled services reduce time-to-value. That means faster discovery, quicker prototypes that become production-ready, and fewer cycles of rework.
If the venture can demonstrate these elements early, it can earn trust quickly—even in a market dominated by established players.
The role of venture backing: why Mayfield and Aramco Ventures matter
Venture capital in enterprise software and services often signals one of two things: either the company is building a platform that scales, or it’s building a delivery model that can be replicated. Mayfield’s involvement suggests the former is likely. Mayfield tends to back companies that can grow beyond services headcount and become technology-driven.
Aramco Ventures adds a different kind of signal. Strategic investors often push for real-world applicability and operational rigor. Industrial and energy enterprises have demanding requirements: uptime, safety, compliance, and integration with legacy systems. If the startup can prove value in such environments, it can translate that credibility to other regulated industries.
In other words, the backing may not just fund growth—it may shape the venture’s priorities toward enterprise-grade execution rather than experimental pilots.
A potential wedge: AI-enabled enterprise transformation
The most compelling wedge for a new IT services challenger is not generic “digital transformation.” It’s transformation that is tightly coupled to enterprise software and measurable operational improvements.
Consider common enterprise pain points:
Manual reconciliation between systems
Finance and operations teams often spend significant time reconciling data across ERP, spreadsheets, and downstream systems.
Slow incident resolution
When issues occur, teams rely on tribal knowledge and scattered documentation. AI can help retrieve relevant context and propose remediation steps, but only if it’s integrated with the organization’s actual knowledge sources.
Inefficient development and testing
Large enterprises have complex codebases and heavy compliance requirements. AI can accelerate coding and testing, but only if it’s connected to the development lifecycle and validated properly.
Process bottlenecks
Supply chain, procurement, and order management processes can be slow due to approvals, exceptions, and inconsistent data.
A startup that combines SAP and AI talent could target these areas with a delivery model that uses AI to reduce cycle times while improving consistency. The key is that the AI must be grounded in enterprise context—process definitions, system constraints,