Affinda launches agentic AI platform for any document workflow

Affinda has launched a new agentic AI document processing platform aimed at enabling enterprises to automate any document workflow with more than 99% accuracy in extracting and structuring data.

According to the company, past efforts to implement AI in enterprise document processing have been slowed by high costs, lengthy delays, and solutions designed mainly for repetitive formats such as invoices.

Existing technologies have not offered a single system that balances ease of set-up, accuracy, and the ability to handle any document type in an end-to-end workflow.

The new platform uses agentic AI and persistent model memory, which Affinda states allows for fast configuration, high accuracy, and flexibility to extract data from any document in any format.

The system combines large language models (LLMs) with retrieval-augmented generation (RAG) to increase accuracy by applying the appropriate context during processing.

Anthony England, Head of Growth at Affinda, said: "Until now, enterprises that wanted to automate their document-heavy processes were forced to compromise on accuracy, time to value or flexibility."

"Now they can get all three. Affinda's platform can accurately extract any information from any document, in any format, fast. It's world class, and our customers are blown away by how quickly they can get set up and how powerful the platform is."

England also stated, "Using the Affinda platform, organizations can now build and customize document processing models themselves in just a few clicks, paying only for what they use."

"By removing the need for extensive DIY and in-house R&D, we're taking the guesswork out of automating document workflows, and we're giving time back to organizations to focus on core projects and innovation."

The system is delivered through custom solutions or self-service and API integrations.

It supports 56 languages and interfaces with more than 400 enterprise systems, which Affinda says enables organisations in any sector or geography to augment their existing infrastructure instead of replacing it.

The company claims to have addressed a significant challenge in document automation: teaching machines to read documents in a way that mimics human understanding.

Unlike older machine learning solutions which required prolonged training and ongoing maintenance, Affinda's new platform reportedly learns immediately from every document and user instruction. In contrast to LLMs used in isolation, Affinda's offering brings together RAG model memory, agentic workflows, proprietary algorithms for reading order, and optical character recognition tools.

With the new platform, onboarding that would have previously taken months can now be accomplished in minutes, according to Affinda. Pricing is offered according to usage, allowing organisations to begin on a small scale during experiments, and then expand as needed.

The platform is available on a free trial basis covering all features, designed to let users build practical models before any financial commitment.

Andrew Bird, Head of AI at Affinda, said: "At Affinda, we empower organizations to combine the best of artificial and human intelligence. Built with an agentic AI approach, our new document processing platform sets a new benchmark – pairing large language models with retrieval-augmented generation – a method that boosts accuracy by adding the right context."

"The result is straight-through processing at scale, giving businesses powerful gains in productivity and efficiency."

Affinda's platform has been implemented in several organisations to handle heavy document processing workloads. StateCover Mutual, an insurance provider, processes over 300,000 case management documents each year across 80 document types using the technology.

Nick Tran, Business Analyst at StateCover Mutual, said: "Affinda's AI platform has allowed us to process over 300,000 documents annually across 80 different document types with superior accuracy and efficiency. This transformation means our people can spend less time on administration and more time supporting our members."

Northline, a freight management and logistics business, uses Affinda's platform to automate the handling of a range of logistics documents.

Jorg Both, Head of Business Systems at Northline, said: "We are now processing over 120,000 Proof-of-Delivery, Pallet and Manual Charge documents annually with an efficiency and accuracy that simply wasn't possible before. Automating validation and document handling across all 13 depots has reduced manual effort, errors and delays – ultimately allowing us to deliver better service to our customers.

Both added, "Affinda's AI document processing platform has streamlined workflows across all 13 depots. Achieving 82% straight-through processing from the start has given us a strong platform to extend automation into invoices and dangerous good notifications linked to Connotes. We've seen significant productivity gains that support our revenue by freeing up our teams to work on higher value work. It's been a pleasure working with the Affinda team and not only have we been impressed by Affinda's platform, the fast ROI and efficiency gains we've experienced, the customer support has been excellent."

Affinda says its platform can be customised to handle any type and format of document workflow, with users able to create bespoke models quickly and deploy them across a wide range of business scenarios.