Novel Patterns Wins 'Best Agentic AI Application' at ET Enterprise AI Awards 2026 for Bringing Governed AI Reasoning to Lending

The recognition acknowledges Novel Patterns’ application of agentic AI to complex lending decisions through DecisionOS, its decision intelligence Platform for financial institutions. DecisionOS connects financial evidence, policy, reasoning and workflow to help AI move beyond task automation and participate intelligently across the lending lifecycle. Novel Patterns’ solutions are currently live with 140+ financial institutions across India, Africa and Southeast Asia.

Ashwani Arora, CEO, Novel Patterns, said, “Banks and NBFCs don’t adopt AI simply because it is fast. They adopt it when it can improve decision capacity while remaining accurate, explainable and defensible in front of a regulator, an auditor and a credit committee. We built DecisionOS around a simple discipline: AI reasons on evidence, operates within policy, and every output carries a traceable record of how it got there.”

DecisionOS: Intelligence Across the Lending Lifecycle

• CRM Intelligence - every lead captured, qualified and intelligently routed.

• Journey Intelligence - onboarding journeys that dynamically adapt to customer context.

• Credit Intelligence - bank statements, financial statements, bureau, tax and salary data transformed into a decision-ready view, with reasoning attached.

• Document Intelligence - documents classified, extracted and checked for inconsistencies and potential tampering.

• Policy Intelligence - policy intent and circulars converted into executable, versioned and simulated rules.

• Post-Disbursal Intelligence - financial and behavioural signals monitored to identify emerging borrower stress before missed payments become the first signal.

• Incentive Intelligence - payouts and reconciliation orchestrated across channels.

• Collateral Management Intelligence - asset value, ownership and exposure monitored throughout the loan lifecycle.

• Partner Portal Intelligence - partner ecosystems onboarded, allocated and monitored through a common intelligence layer.

And many more intelligence capabilities across the financial lifecycle.

Together, these capabilities connect customer context, financial evidence, policy, reasoning and workflow into a common decision fabric.

Built for Financial Intelligence - Not Generic AI

The intelligence behind DecisionOS is purpose-built for financial decision-making. The platform is designed around a multi-layer trust architecture incorporating source-grounded processing, policy and rule validation, cross-document reconciliation, confidence scoring and human-in-the-loop review for exceptions.

 It analyses more than 400 financial and behavioural indicators and bring more than 1,000 behavioural risk signals into credit assessment. Proprietary financial ontologies provide the domain context required to interpret relationships across different sources of financial information.

DecisionOS combines this intelligence with policy and workflow, helping institutions understand what the evidence means, which policy applies, why a recommendation was generated and what needs to happen next.

From AI Automation to Decision Intelligence

For Novel Patterns, the award is a milestone in a larger ambition: to build a connected intelligence layer for financial decision-making.

The objective is not to replace human judgement, but to let AI handle more of the complexity while people retain authority where judgement matters - with consequential decisions remaining explainable and auditable.

Intelligence that reasons. Decisions that remain human. Outcomes that can be trusted.

About Novel Patterns

Novel Patterns is an AI-native financial intelligence company building intelligent decision infrastructure for financial institutions. Its platforms combine financial domain intelligence, AI reasoning, decision orchestration and governance across lending and investment management. Novel Patterns’ solutions are used by 140+ financial institutions across India, Africa and Southeast Asia.