Hexaware Technologies has introduced Tensai for Reasoning Ops, an enhancement to its Tensai Agentic ITOps platform. The latest stage of the platform combines operational signals with enterprise context to provide evidence-backed action recommendations. Human experts validate and execute these recommendations, supporting a shift toward operations that focus on reducing recurring demand rather than only resolving incidents more quickly.
Traditionally, enterprises have relied on scripts and runbooks to automate operational tasks. While these approaches improve task execution, they are limited in their ability to reason, interpret context, and connect information across different systems. This can contribute to fragmented tooling and siloed knowledge, creating gaps in system-level intelligence. Tensai for Reasoning Ops addresses this by using operational context from observability data, configuration management databases (CMDBs), topology, change records, and dependency data. The platform’s agents analyse signals, context, and policies to support decision-making beyond scripted automation.
Key Features:
Evidence-based Recommendations: Each recommended action is supported by relevant operational signals and contextual information.
Policy Verification: Proposed decisions are assessed against defined risks and policies before human validation.
Cross-tower Reasoning: Agents analyse information across operational domains that are often managed separately.
Self-improvement Mechanism: Outcomes are continuously evaluated to improve future recommendations and decision-making.
The platform represents a step in Hexaware’s transition from Traditional Ops to Preventive Ops. Initial outcomes indicate improvements in operational efficiency, service quality, and demand reduction.
Measured against each client’s pre-AI baseline, the platform is designed to target a reduction in manual intervention of 35–45%, a 10–18% decrease in cost-to-serve, and a 25–40% improvement in mean time to recovery (MTTR). It also aims to improve service-level agreement (SLA) reliability by 10–20% and reduce incident demand by 5–12%. These metrics will be tailored to individual environments and assessed through the Tensai Customer Value Scorecard, supporting adoption across mid-to-large enterprises.
As organisations continue to integrate AI into IT operations, the focus is expanding from faster incident resolution toward identifying and reducing predictable sources of demand. Tensai for Reasoning Ops supports this approach by helping address the underlying causes of operational issues proactively, rather than responding only after incidents occur, contributing to the development of preventive and increasingly autonomous operations.
