Jitterbit unveils Harmony upgrade with layered accountable AI agents

Jitterbit has announced the expansion of its Harmony platform with accountable, layered AI technology and the addition of enterprise-ready AI agents to its product portfolio.

The company's approach centres on providing not merely AI assistants to assist users, but AI agents capable of autonomously performing complex tasks on behalf of enterprises. A layered AI architecture underpins Harmony, which allows customers to engage with "agentic AI" in three distinct ways: creating their own agents using low-code or natural language, accessing pre-built agents from a curated marketplace, or working with Jitterbit's Professional Services team to develop custom agents according to their requirements.

Jitterbit President and Chief Executive Officer Bill Conner said, "We're not just automating; we're transforming how enterprises operate. Jitterbit is delivering the first layered AI and low-code architecture to democratise end-to-end automation with a focus on power, efficiency, and AI accountability. This isn't just about automating tasks; it's about architecting intelligent, autonomous agents with a unified platform that eliminates the 'data divide' between enterprise data and applications."

The Harmony platform, which includes integration Platform as a Service (iPaaS), App Builder, API Manager, and Electronic Data Interchange (EDI) solutions, is designed to bring together line-of-business leaders and IT or Information Systems experts. This collaborative environment enables both groups to develop AI agents that can be integrated with existing enterprise architecture, thereby seeking to enhance operational efficiency while ensuring control, transparency, and accountability across business processes.

CTO Manoj Chaudhary highlighted the flexibility of the platform's layered AI capabilities. "The beauty of our layered AI approach is that our customers can use their current investments to design and implement AI agents, or have Jitterbit do it for them," he said. "We're not isolating AI to a particular product or feature; customers have full control to use low-code or natural language to take their existing implementations and quickly design new AI agents to accelerate their current systems and processes in ways they've never imagined."

Jitterbit's solution is designed with security, governance and accountability at its core. The platform allows organisations to create agents with natural language or low-code instructions, access a growing marketplace of vetted AI agents (including third-party offerings), or fully outsource custom agent development to Jitterbit's professional services team. This service will be available to customers from May 2025.

Chaudhary stated, "Regardless of how AI agents are built and deployed, trust and accountability are Jitterbit's core tenets. We're empowering organisations with the 'checks and balances' to ensure agents are not only making correct logical decisions, but also providing guardrails to mitigate issues like toxicity and AI hallucination. And, as always, we're providing mechanisms for human oversight and verification for extra layers of accountability."

Jitterbit's recent study, 'The 2025 Automation Benchmark Report: Insights from IT Leaders on Enterprise Automation & the Future of AI-Driven Businesses', reported that 99% of surveyed enterprises have already integrated AI into their operations. The survey also found that 31% of enterprises are planning for agentic AI, which points to growing interest in more autonomous decision-making solutions and the need for layered AI alongside comprehensive end-to-end automation.

Richard Guest, EMEA Delivery Director at Jeld-Wen, commented on the importance of the new technology. "Accountability is no longer a 'nice-to-have' but a critical driver of business value in the age of agentic AI. A focus on layered and accountable AI enables organisations to confidently scale their automation initiatives, knowing they have the control and visibility needed to achieve strategic outcomes."

Potential applications of agentic AI include customer service agents, supply-chain automation, HR onboarding, planning for sales accounts, legal research, and financial analysis. The technology is intended for adoption across a range of sectors, reflecting the growing role of autonomous AI agents in streamlining business processes.

In addition to launching the layered AI architecture, Jitterbit announced the general availability of AI assistants for its App Builder and API Manager from June 2025. The App Builder AI Assistant will allow users to build or modify applications through natural language interactions and an interface that can be customised by uploading reference images. The API Manager AI Assistant is intended to reduce API development times by enabling users to build APIs more efficiently with the support of AI integration.