Real-time tax compliance puts agentic AI accuracy to the test

Avalara Inc. is applying agentic AI to its tax compliance platform, balancing autonomy with the penny-perfect accuracy required for financial systems, according to Marcus Larner, senior vice president of engineering at Avalara. The company's Avi agent, launched November 3, 2025, uses a deterministic tax engine drawing from more than 87,000 regulatory sources across more than 190 countries, and its AvaTax product is deployed in Google Cloud data centers to support Shopify's workloads. Larner discussed these efforts during the Google Cloud AI Agents in Action Series on theCUBE.

Real-time tax compliance puts agentic AI accuracy to the test AI-powered tax compliance has to meet a standard that many artificial intelligence applications don’t: The answers must be exactly right. While large language models can generate unpredictable results, tax calculations require accuracy, speed and reliability across thousands of jurisdictions. That tension has shaped the way Avalara Inc. applies agentic AI to its transactional tax and compliance platform https://www.avalara.com/us/en/index.html . The company combines agents that interpret requests and orchestrate workflows with a deterministic tax engine designed to produce precise calculations. “When you talk about tax in any financial system, really, you have to be penny-perfect,” said Marcus Larner https://www.linkedin.com/in/marcus-larner/ pictured , senior vice president of engineering at Avalara. “You have to have a high degree of trust. You have to be highly performant and highly resilient. Those are things you don’t typically associate with AI.” Larner spoke with theCUBE’s John Furrier https://www.linkedin.com/in/furrier/ during the Google Cloud AI Agents in Action Series https://www.thecube.net/events/google/ai-agents-in-action on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how Avalara is balancing autonomy with accuracy and using Google Cloud technologies to extend its agentic tax capabilities. Disclosure below. AI-powered tax compliance balances autonomy and accuracy Avalara has used AI and machine learning for more than a decade, giving the company experience in determining which parts of a tax workflow are suited to AI. Its AI-powered tax compliance strategy is to apply agents selectively while preserving the performance and accuracy required of financial systems, according to Larner. “We have to balance where to apply AI smartly and intelligently so that we maintain that high performance, that high degree of trust and that high degree of accuracy,” he said. “With Google’s partnership, that’s … how we’ve achieved that.” Avalara’s Avi https://newsroom.avalara.com/2025-11-03-Avalara-Launches-Avi-and-a-Network-of-Secure-Compliance-Agents-Built-for-Action,-Designed-for-Human-Oversight , an agent that provides a natural-language entry point into the company’s compliance products, reflects the company’s AI-powered tax compliance approach. During a live demonstration, Avi validated an address, identified a tax code, calculated tax, mapped tax liabilities and generated a transaction report. Its tax engine draws from more than 87,000 regulatory sources across more than 190 countries, according to Avalara. “The Avi agent in the front end is receiving requests,” Larner explained. “It is deciding, ‘OK, which piece of the compliance puzzle needs to solve this? What data do I need to bring in? What source of truth and groundings do I need to apply in order to have a highly accurate solution?’” Avalara’s approach to resiliency also extends to where its services operate. Its active-active strategy distributes its AvaTax https://www.avalara.com/us/en/products/calculations.html product across regions and cloud providers, allowing customers to switch locations when disruptions occur. That architecture allowed Avalara to place AvaTax alongside Shopify’s workloads in Google Cloud data centers, according to Larner. “We partnered with Google to apply and deploy our solutions globally, and that enables us to be where our customers want us to be,” he said. “When Shopify rolled out AvaTax, they needed us to be exactly in the same physical location, physical data centers that they were in, in order to have that high performance and high resiliency.” Real-time demands reshape tax and finance The need for trusted AI-powered tax compliance is increasing as governments move from periodic tax reporting toward real-time integrations, including electronic invoicing. At the same time, lower compliance thresholds are raising audit risks and adding complexity for companies operating across jurisdictions, Larner explained. “We have to become a lot more accurate and build and maintain that trust with companies,” he said. “Within Avalara and within businesses, AI is providing that simplification: Agents are increasingly able to take on complex workflows and free up the people and the human resources to do the complex tasks. We’re simplifying workflows, we’re making things more streamlined, and we have a changing role of finance.” Access to tax information in near real time could also allow finance teams to use it for more than retrospective reporting. Tax obligations can become part of forecasting, product decisions and broader business planning, Larner added. “Finance now becomes a strategic player,” he said. “It’s not just a reporting mechanism, but how can you start to forecast and model and use all of the data and AI to strategize and mold your business?” Google Cloud also supports the ecosystem behind Avalara’s AI-powered tax compliance strategy. The Agent Development Kit https://adk.dev/ provides a framework for building agents, while the Agent2Agent protocol https://cloud.google.com/blog/products/ai-machine-learning/agent2agent-protocol-is-getting-an-upgrade allows agents built on different systems to communicate and coordinate tasks. Because Avalara makes Avi available through Google Cloud Marketplace, maintaining security controls https://siliconangle.com/2026/04/22/google-cloud-next-new-security-operations-agents-wiz-integrations-agent-governance-tools/ throughout that expanding agent stack is essential to preserving trust, according to Larner. “The early implementations of these standards really didn’t have the security story nailed down, but now we do,” he said. “Having that infrastructure and the security controls all the way through your stack … we have 100% confidence that our solutions, even with AI and large language models involved, are highly secure and maintain that trust.” Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the Google Cloud AI Agents in Action Series https://www.thecube.net/events/google/ai-agents-in-action : Disclosure: TheCUBE is a paid media partner for the Google Cloud AI Agents in Action Series. Neither Google Cloud, the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE. Photo: SiliconANGLE A message from John Furrier, co-founder of SiliconANGLE: Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network , where technology leaders connect, share intelligence and create opportunities. 15M+ viewers of theCUBE videos , powering conversations across AI, cloud, cybersecurity and more 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network. Are you AWS customer? Support SiliconANGLE Financially by buying your AWS services from our Marketplace portal page and links. About SiliconANGLE Media SiliconANGLE https://cts.businesswire.com/ct/CT?id=smartlink&url=https%3A%2F%2Fsiliconangle.com%2F&esheet=54119777&newsitemid=20240910506833&lan=en-US&anchor=SiliconANGLE&index=9&md5=646b1b564e2259100a2b8638aab0a552 , theCUBE Network https://cts.businesswire.com/ct/CT?id=smartlink&url=https%3A%2F%2Fwww.thecube.net%2F&esheet=54119777&newsitemid=20240910506833&lan=en-US&anchor=theCUBE+Network&index=10&md5=7de2a85f95ab4a4a495cede20b8cb1da , theCUBE Research https://cts.businesswire.com/ct/CT?id=smartlink&url=https%3A%2F%2Fthecuberesearch.com%2F&esheet=54119777&newsitemid=20240910506833&lan=en-US&anchor=theCUBE+Research&index=11&md5=7bb33676722925eb57d588ec343e4f6f , CUBE365 https://cts.businesswire.com/ct/CT?id=smartlink&url=https%3A%2F%2Fwww.cube365.net%2F&esheet=54119777&newsitemid=20240910506833&lan=en-US&anchor=CUBE365&index=12&md5=d310fb35919714e66ad8d42c9c0c1bc6 , theCUBE AI https://cts.businesswire.com/ct/CT?id=smartlink&url=https%3A%2F%2Fwww.thecubeai.com%2F&esheet=54119777&newsitemid=20240910506833&lan=en-US&anchor=theCUBE+AI&index=13&md5=b8b98472f8071b23ebb10ab9a8dd0683 and theCUBE SuperStudios — with flagship locations in Silicon Valley and the New York Stock Exchange — SiliconANGLE Media operates at the intersection of media, technology and AI. Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.