NeoXam launches AI agent platform for investment operations

By Milan Rojan

NeoXam has launched NeoXam Agents, an agentic AI offering designed to help investment firms deploy, manage and govern artificial intelligence across middle- and back-office operations.

The new offering has comprised three elements: NeoXam Agents, a platform for building, running and supervising AI agents; a suite of specialised agents currently in preview; and an early-adopter programme that has been opened to a select group of investment firms from the third quarter of 2026.

Clément Miglietti, Chief Product and Technology Officer at NeoXam, said: “Most of our clients don’t have an AI problem. They have a production problem. They’ve run the pilots and seen the demos. What they can’t do yet is sit an agent in front of a regulator and account for every step it took, so we built that part first.”

NeoXam has said the platform has been integrated directly into its existing software products, enabling AI agents to operate within workflows that manage market and reference data, portfolio and accounting records, reconciliations, compliance monitoring and regulatory reporting. The platform has provided a catalogue of approved agents, audit trails for all actions taken, human approval controls for sensitive decisions and flexibility to switch between underlying AI models.

The company has also introduced a range of specialised agents focused on operational tasks. Within NeoXam Aro, an AI agent has helped automate reconciliation processes, resulting in up to a 75% reduction in manual reconciliation work. In NeoXam PMS, a compliance-focused agent has reduced the time required to address rule-breach alerts by up to 90%, while a configuration agent within DataHub has delivered time savings of around 90% on routine set-up tasks.

Additional agents have been developed for reporting and knowledge management. NeoXam said its reporting agent has assembled client-ready presentations in minutes, while knowledge agents have answered business and technical queries across its product suite to support onboarding and customer service.

Alongside the launch, NeoXam has introduced NeoXam Structure, an AI-powered document processing capability that has extracted information from documents such as broker statements, PDFs and trade confirmations.

The early-adopter programme has been scheduled to run for one year, with participating firms testing deployments during late 2026 and a path to production planned for early 2027.