SlateWorks wins ABBYY hackathon with mortgage auditor
Thu, 6th Aug 2026 (Today)
SlateWorks Global Solutions won ABBYY's hackathon at its Ascend Developer Conference in Bangalore with a mortgage auditing tool called AIM. The competition drew more than 25 teams and produced eight finalists.
The winning entry, ABBYY Intelligent Mortgage Auditor, was built to automate mortgage file reviews by combining document extraction with generative AI and retrieval-augmented generation. Judges said it tackled a manual process in a tightly regulated area of financial services and was ready for deployment.
Event organisers said attendance at the developer conference has quadrupled since the series began four years ago, making the Bangalore gathering the largest so far. Partners, developers and business users from finance, manufacturing, banking, healthcare and retail attended.
As a condition of entry, finalists used ABBYY software with third-party AI tools. Recorded presentations featured teams from Tata Consultancy Services, Proservartner, SlateWorks Global Software, Tangentia, Ellucian and Tech Mahindra.
SlateWorks based its submission on the mortgage industry's high document volumes and compliance burden. In the team's presentation, Sonali of SlateWorks said a typical loan file can run to hundreds of pages and that manual judgment can cause delays and errors.
AIM used ABBYY Vantage to split, classify and extract information from loan files before passing the output to SlateWorks' AI platform. The platform then ran checks including income analysis, employment verification and guideline reviews, while giving auditors access to a chatbot for queries on extracted data.
In the demonstration, the team showed the system flagging failed checks, identifying missing documents and comparing file information with third-party tax transcript data. It said a report for a sample loan was produced in just over three minutes.
Judges asked how failed or missing documents would be handled after review. Sonali said the current version required users to upload documents again and start a new transaction, though the team wanted to add a loop for failed files within the same transaction.
"At SlateWorks Global Solutions, we consistently see the significant effort organisations spend reviewing documents, verifying information and maintaining regulatory compliance - work that is critical yet still largely manual," said Sam Kharidi, Founder and CEO, SlateWorks Global Solutions.
"Winning the ABBYY Ascend DevCon Hackathon validates our vision that combining deep domain expertise with ABBYY's document intelligence and AI can create solutions that extend far beyond traditional document automation," Kharidi said.
Other finalists
Tata Consultancy Services presented Supplier Trust Gateway, a supplier onboarding tool for procurement teams. The team described ABBYY Vantage as an orchestration layer that classified documents, extracted relevant data and routed selected parts of files to different AI models.
The TCS prototype focused on supplier financial statements and registration records. It used one model for financial analysis and another for fraud checks, with staff reviewing the results in a human-in-the-loop process. During questioning, the team said the main benefit was not just speed but the consolidation of work that often sits across finance, compliance and procurement.
Proservartner showed Cash Pilot, a product for accounts receivable and remittance processing. The system connected bank feeds, invoice registers and uploaded remittance files, while a set of software agents handled extraction, exception management and posting workflows.
The presentation stressed that finance specialists often spend more time gathering information from ERP systems, emails and spreadsheets than making decisions. Judges asked whether ABBYY's manual review stage could be embedded directly in the product interface, and presenter Vivek said that was under consideration.
Tangentia took a legal technology approach with Mike, the Law Agent, pitched as a research and review assistant for junior lawyers. Presenter Inacio Fernandez said it could process court and contract documents and connect the extracted content to legal research tools, helping users identify concerns, missing clauses and proposed redlines.
Ellucian also focused on legal work, this time for Indian advocates managing clients, case files and court records. Its team said lawyers often handle large volumes of documents and risk missing clauses or evidence, so it built custom OCR skills for common legal scenarios and connected the output to a vector database and AI search functions.
Tech Mahindra presented AVA, a claims adjudication assistant for healthcare. The system processed incoming claims, scored fraud risk and routed difficult cases for further review, while using voice and chat features to help adjudicators query results and inconsistencies.
Broader shift
The hackathon prize included USD $1,500 in Amazon vouchers, automatic entry into ABBYY's 2027 Most Valued Partner programme and fast-track access to its Innovation Hub. Sarita Behera, who led hackathon logistics and the MVP community programme, said 28 teams entered and that finalists built complete working solutions.
ABBYY presented the contest as evidence that customers and partners are moving from AI trials to more targeted operational tools. The projects reflected that shift, with most centred on document-heavy workflows in regulated or administratively complex sectors.
"The conversation has shifted, and organisations are moving beyond AI experimentation and demanding solutions that are accurate, auditable and ready for production," said Neil Murphy, Chief Revenue Officer, ABBYY.