Rivvun AI Inc. today announced a $7.55 million oversubscribed seed round led by Sitara Capital and 3one4 Capital, to deploy an autonomous AI execution layer purpose-built for enterprise spend and revenue recovery.
The scale of the problem is staggering. McKinsey research finds that enterprise procurement functions lose up to one-third of planned savings during execution — with an additional 3– 4% of total external spend lost to transaction inefficiency and noncompliance. Across fortune 2000 revenues that compounds to more than $2T in value that never reaches the bottom line. The money isn’t lost to fraud or bad contracts. It disappears in the gap between what was contractually committed and what enterprise systems were ever built to collect.
Built by the Executives Who Saw This Problem at Scale
Anand Veerkar and Niranjan Umarane spent the last decade as senior executives at Icertis, where they helped scale the company to more than $350 million ARR and built a platform governing some of the world’s largest commercial portfolios. Across every industry, the pattern was consistent: terms of trade were precisely structured; financial execution against them was not. Money owed under negotiated agreements quietly went uncollected — not because anyone decided to leave it, but because no system in the enterprise stack was designed to recover it. They left to build that system. They are joined by serial entrepreneur Patrick Linton, who brings deep experience scaling global operations for enterprise software companies.
The Problem Is Structural. So Is the Solution.
ERP systems record transactions. CRM tools track relationships. Procurement platforms manage approvals. None of them enforce outcomes. Rivvun’s autonomous AI execution layer connects to existing ERP, CRM, and procurement systems, interprets commercial obligations, identifies what hasn’t settled as agreed, and initiates recovery at the transaction level. No rip-and-replace. No new system of record.
Two agentic families power the platform: Spend Assurance on the buy side — recovering supplier rebates, pricing commitments, and procurement obligations that have gone unenforced; and Margin Defense on the sell side — recovering customer settlement variances, trade term discrepancies, and revenue that left the P&L without authorization.
Built Vertical-First, Because Leakage Isn’t Generic
Chargeback mechanics in pharma — GPO compliance, government pricing obligations — look nothing like settlement gaps in banking or trade term failures in CPG. Generic AI produces generic results. Rivvun deploys with vertical-specific agent logic tuned to the precise failure patterns of each industry, across Pharma, Healthcare, Banking, CPG/Retail and Industrial
Anand Veerkar, CEO and Co-Founder, Rivvun AI commented: “The enterprise has spent years being told AI will transform how it operates. What it needed was AI that creates direct, measurable impact on the P&L – not productivity narratives, not dashboards. Rivvun closes the gap between what was agreed and what was collected, recovering money that goes straight to the bottom line.
Sachin Bhanot, Managing Partner, Sitara Capital added: “We’ve invested in enterprise technology for years. The winners tie their value directly to a number the CFO can see on the P&L. Rivvun does exactly that with precision rare for a company at this stage – and with a founding team that has already built a category-leader in this space.”
Anurag Ramdasan, Partner, 3one4 Capital said: “The team at Rivvun is one of the strongest founder-market fit we’ve seen in the vertical AI category so far. They are not pitching a horizontal AI solution and hoping for enterprises to extract value out of it. They are delivering ROI on AI for large enterprises from the first day of implementation, which is very critical for enterprise AI adoption. This rigor comes from the deep expertise of the founders, and we are incredibly excited to back such a transformational team at seed stage.”
To participate in our interviews, please write to our CyberTech Media Room at sales@intentamplify.com
