Get paid faster: How Intuit’s new AI agents help businesses get funds up to 5 days faster and save 12 hours a month with autonomous workflows
Become a member of the event that is trusted by business leaders for almost two decades. VB Transform brings together the people who build the real Enterprise AI strategy. Leather
Intent Has been traveling in recent years with Generative AI, where technology is included as part of its services at Quickbooks, Credit Karma, TurboTax and Mailchimp.
Nowadays the company is taking the next step with a series of AI agents who go further to transform how small and medium-sized companies work. These new agents work as a virtual team that automates workflows and offers real -time business insights. They include opportunities for payments, accounts and finances that directly affect the business activities. According to Intuit, customers save up to 12 hours a month and are paid on average up to five days faster thanks to the new agents.
“If you look at the process of our AI experiences at Intuit in the early years, AI was built into the background, and with Intuit Assist you saw a shift to give information back to the customer,” Ashok Srivastava, Chief Ai and Data Officer told Venturebeat. “What you see is a complete redesign. The agents actually do work on behalf of the customer, with their permission.”
Technical architecture: from start kit to production agents
Intuit has been working on the path from assistants to Agentic AI for some time.
In September 2024, the company described its plans to use AI to automate complex tasks. It is an approach that is firmly built on the Generative AI control system (Genos) platform of the company, the basis of its AI efforts.
Earlier this month, Intoge announced a series of efforts that further expand its possibilities. The company has developed its own fast optimization service that will optimize questions for every large language model (LLM). It has also developed what it calls an intelligent data cognition layer for company data that can understand various data sources that are needed for Enterpriseworkflows.
To one step further, Intuit developed an agent Starter Kit who builds on the technical basis of the company to make Agentic AI development possible.
The Agent Portfolio: From cash flow to customer management
With the technical basis, including agent Starterskits, intuitively has built up a series of new agents who help business owners get things done.
Intuit’s Agent Suite demonstrates the technical refinement needed to transfer from predictive AI to autonomous workflow version. Each agent coordinates prediction, natural language processing (NLP) and autonomous decision -making within full business processes. They include:
Payment agent: Optimizes autonomous cash flow by predicting late payments, generating invoices and performing follow -up series.
Accounting: Represents the evolution of intuit, from rules -based systems to autonomous accounting. The agent is now handling autonomously transaction categorization, reconciliation and completion of the workflow, which supplies cleaner and more accurate books.
Financial agent: Automates strategic analysis that is traditionally needed to require special Business Intelligence (BI) tools and human analysts. Offers key Performance Indicator (KPI) analysis, scenario planning and prediction based on how the company is doing against peer -benchmarks and at the same time generating growth recommendations.
Intuit also develops customer hub agents who help with the acquisition tasks of customers. Payroll processing and efforts for project management are also part of the future release plans.
Beyond Conversational UI: Task -oriented agent design
The new agents mark an evolution in how AI is presented to users.
Intuit’s interface redesign reveals important user experience principles for the implementation of the Enterprise agent. Instead of stimulating AI possibilities to existing software, the company has fundamentally restructured the QuickBooks user experience.
“The user interface is now really focused on the business tasks that must be performed,” explained Srivastava. “This allows real -time insights and recommendations to come directly to the user.”
This task-oriented approach is in contrast with the chat-based interfaces that dominate the current Enterprise AI tools. Instead of demanding users that they learn prompts strategies or navigate conversation flows, the agents work within existing business workflows. The system includes what intuitively calls a “business feed” that contextually activating actions and recommendations pops up.
Trust and verification: The challenge with closed loop
One of the most important aspects of the implementation of Intuit takes on a crucial challenge in the use of autonomous agent: verification and trust. Enterprise AI teams often struggle with the Black Box problem -How do you ensure that AI agents perform correctly when they work autonomously?
“To build trust with artificial intelligence systems, we must offer the customer a points of proof that what they think is happening,” Srivastava emphasized. “That closed loop is very, very important.”
The Intuit solution includes the right to build verification possibilities in Genos, so that the system can provide evidence of agent actions and results. For the payment agent, this means that users are shown that invoices have been sent, keeping delivery and demonstrating the improvement in payment cycles that is the result of the agent’s actions.
This verification approach offers a template for business teams that use autonomous agents in business processes with high deployment. Instead of asking users to trust AI outputs, the system offers auditable paths and measurable results.
What this means for companies looking for agentic AI
Intuit’s Evolution offers a concrete route map for business teams that plan autonomous AI implementations:
Focus on the completion of the workflow, no conversation: Target specific business processes for end-to-end automation instead of building general chat interfaces.
Build Agent Orchestration Infrastructure: Invest in platforms that coordinate forecast, language processing and autonomous performance within uniform workflows, uniled AI tools.
Design verification systems in advance: Take out comprehensive audit paths, follow outing and user reports as core options instead of afterwards.
Map Workflows Before building technology: Use customer advice programs to define agent options based on actual operational challenges.
Re -design plan for interface: Optimize UX for agent-driven workflows instead of traditional software navigation patterns.
“As large language models are produced, the experiences built on it are much more important,” said Srivastava.
Source link
