StitcherAI has introduced a new IT finance intelligence platform designed to steer every technology investment decision—whether by humans or autonomous agents—against business-aligned financial context in real time. The company, led by founder and CEO Udam Dewaraja, a former global head of IT finance at Citi, aims to solve a persistent problem: the disconnect between rapid AI and cloud spending and the slower, reactive financial oversight mechanisms that govern them.
FOCUS First: A Semantic Engine for Unified Cost Understanding
At the heart of StitcherAI's platform lies a semantic engine built on FOCUS—the FinOps Open Cost & Usage Specification. FOCUS is an open billing standard co-created by Dewaraja and adopted by major cloud providers including AWS, Azure, and Google. This engine unifies data from cloud, AI, software-as-a-service (SaaS), and other vendor sources, modeling costs in business terms that resonate with boards: products, customer segments, margins, unit-economics KPIs, forecasts, and cost controls. The resulting intelligence flows continuously into an organization's existing data lakes, business intelligence platforms, JIRA, Slack, and enterprise resource planning systems, eliminating the need for manual tagging or dashboard checks.
The FOCUS standard itself represents a significant industry effort to normalize billing datasets across technology vendors. By leveraging it, StitcherAI ensures that enterprises can compare costs across disparate environments without facing the complexity of multiple pricing models. This is especially critical given the rise of agentic workflows, where autonomous agents may interact with dozens of tools and vendor APIs, each with its own pricing structure and token math. Without a unified semantic layer, finance teams are left trying to piece together costs from fragmented data sources, a process that can take months and often results in incomplete insights.
Omnipresent Reasoning Engine: Steering Investment Decisions in Real Time
The platform's second key component is an omnipresent reasoning engine comprising specialized agents that inject organization-specific financial context into agentic workflows and AI platforms such as Claude, Cursor, and Codex. As IT dollars are committed—whether through human decision or automated agent action—the engine ensures that each spend is evaluated against predefined financial goals and constraints. This shift from managing cost after the invoice arrives to making context-aware decisions before money leaves the organization transforms IT finance from a reactive function to a proactive one. StitcherAI claims it can cut the time required to understand business impact from months to near-instantaneous.
Traditional IT finance tools rely on engineers remembering to tag resources, finance teams manually policing spend, and stakeholders periodically checking dashboards. At human scale, this approach already generates friction; at agent scale, where thousands of autonomous decisions can occur per second, it becomes completely unworkable. The reasoning engine addresses this by embedding financial guardrails directly into the decision-making loop, enabling engineers and agents to operate with full awareness of budget constraints, unit economics, and ROI expectations.
Human Scale, Agent Scale: The Growing Imperative
Dewaraja observed this problem firsthand while leading global IT finance at Citi. He noted that both humans and agents routinely failed to check existing dashboards when committing spend, often lacking access to all relevant data. With StitcherAI, the goal is to reimagine how organizations manage modern IT spend using AI to automate low-impact decisions while providing business-aligned financial intelligence for higher-order investment decisions. This dual approach acknowledges that not every cost decision requires human intervention—automation can handle routine approvals based on predefined rules—but strategic investments still demand human judgment informed by comprehensive financial context.
Research from IBM underscores the urgency: only 25% of AI initiatives have delivered expected return on investment over the past few years, and just 16% have scaled enterprise-wide. Engineers and autonomous agents using tools like Claude Code, Cursor, and Codex now make architecture and infrastructure decisions in seconds, while the IT finance infrastructure governing them was built for weekly review cycles. This mismatch creates a pattern where AI and IT spend is committed with no consideration of ROI, forcing finance teams to firefight when invoices arrive. StitcherAI aims to close this loop by ensuring that every spending decision—whether by a human developer or an agentic workflow—is made with real-time financial context.
Unintended Execution Paths: The Risk of Agentic Drift
The problem intensifies as agentic workflows drift onto unintended execution paths or hallucinate in ways that original designs never anticipated. Fortune 100 enterprises report spending 18 to 24 months and millions of dollars in labor to build in-house solutions for wrangling IT spend. Yet such bespoke approaches are often too slow and expensive for most organizations now facing rising AI and IT investments. StitcherAI offers a faster, more scalable alternative by embedding financial intelligence directly into the technology stack, allowing enterprises to avoid costly rework and align IT spending with business objectives from the outset.
Tim Crawford, CIO Strategic Advisor at AVOA and a member of The Wall Street Journal Technology Council, emphasizes that CIOs and CFOs are no longer asking how to manage cloud and AI costs. Instead, they are asking which AI investments are producing returns, how to scale those winners, and how to quickly kill underperforming initiatives. What they need, Crawford says, is a system that puts the enterprise's financial context into the workflow at the point of decision, for both humans and agents. StitcherAI's platform appears designed to meet that exact need, embedding financial guardrails directly into the decision-making fabric of the organization.
The company's approach also addresses the challenge of cost attribution in agentic environments. When an autonomous agent invokes multiple APIs, runs different models, and uses various cloud resources to complete a task, the unit cost can vary dramatically from request to request. Without a system that tracks these costs in real time and maps them to business outcomes, it becomes nearly impossible to assess ROI or optimize spending. StitcherAI's semantic engine solves this by normalizing all spend data against the FOCUS standard and then applying business context—such as product line, customer segment, or strategic initiative—to every transaction. This allows finance teams to see exactly where value is being created and where waste is occurring.
As enterprises continue to accelerate their AI adoption, the need for robust IT finance intelligence will only grow. StitcherAI's platform represents a significant step forward, offering a system of intelligence that can keep pace with the speed of modern technology investments. By bridging the gap between rapid, agent-driven spending and slower, traditional financial oversight, it enables organizations to make smarter, more profitable decisions about their AI and cloud investments.
Source: Computerweekly News