Gartner has published its inaugural Emerging Market Quadrant for No-Code Agent Builders — a brand new analyst category that recognizes the platforms shaping how enterprises build and deploy AI agents. Tray.ai has been named a Pioneer. We’re proud of the recognition, and we think what it reflects about this market is worth explaining.
Last year, just 17% of enterprises had AI agents running in production. This year, 42% plan to deploy them. That gap — from intent to reality, at scale, across a real enterprise environment — is not a technology problem. The models are capable. The interfaces are accessible. The demos are impressive.
The gap is an orchestration problem. And it has been hiding in plain sight.
Where enterprise AI programs break down
When enterprise AI agent initiatives stall, the post-mortems tend to surface the same findings: the agent worked in the pilot. It connected to the right data, executed the right task, produced a usable output. Then someone tried to run it across three business units, connect it to a legacy ERP, apply role-based access controls, and give a non-technical operations manager the ability to modify it without filing an IT ticket. At that point, the initiative stopped.
This is the orchestration layer. It is not glamorous. It rarely appears in vendor keynotes. But it is where enterprise AI agent programs succeed or fail in practice — in the governance, the multi-system connectivity, the security model, and the infrastructure that makes an agent reliable and auditable at scale rather than impressive in a sandbox.
This is the problem Tray.ai was built to solve. Not as a response to the AI agent wave, but as the foundation underneath it. Merlin Agent Builder shipped in December 2024 — before “agentic AI” had become a fixture in analyst reports or a competitive talking point for the integration platform market. That timing matters, because the foundation Tray.ai was laying then is exactly the infrastructure enterprises are now discovering they need.
What “Pioneer” actually means in this context
The 2026 Gartner Emerging Market Quadrant for No-Code Agent Builders includes organizations whose names signal how seriously this market is being taken: Microsoft, Google, and AWS appear in the quadrant’s Market Shapers category. That context matters. When hyperscalers are present in an emerging market quadrant, analysts have concluded this is not a speculative category — it is infrastructure. The question is no longer whether enterprises will run AI agents. It is which architectural bets will hold up when they do.
Gartner uses the Pioneer designation to identify vendors with strong innovation and forward-looking product direction who are building before the market has fully formed around them. It is a position of conviction — companies that are shaping a market rather than responding to one.
Being named a Pioneer alongside vendors who are pivoting deterministic automation and integration platforms toward agentic AI is telling. The distinction matters to enterprise buyers: there is a meaningful difference between a company that built orchestration infrastructure over a decade and a company that surfaced an “agentic” layer on top of an existing product in the last twelve months. Both may appear in the same quadrant. They do not represent the same architectural bet.
Tray.ai’s Pioneer position reflects a decade of building for enterprise integration complexity — 700+ connectors, infrastructure engineered for multi-system data flows, and a governance model designed for enterprise security requirements before the phrase “AI agent” was in common use. The Pioneer label does not describe where Tray.ai is going. It describes what Tray.ai has already built.
The stakes: what 2028 looks like if you plan correctly now
Gartner’s forward projection for this market is significant: by 2028, citizen developers are expected to build and maintain more AI agents than traditional developers build apps, bots, and workflows combined. That is not a prediction about technology. It is a prediction about organizational structure and who owns automation inside the enterprise.
For IT and architecture leaders, the planning implication is immediate. The question is not whether your organization will have non-technical builders deploying AI agents. It is whether your infrastructure will be ready to govern what they build.
This is the specific problem Merlin Agent Builder addresses. It is a no-code environment that lets operations leads, analysts, and department managers build and deploy production AI agents — without writing code, and without creating governance debt for IT. The agent runs on Tray.ai’s enterprise infrastructure. The oversight stays with the teams who need it.
The companion problem is what happens at the enterprise perimeter: how do AI agents, built by different teams, using different models, connect securely to internal systems and third-party services without creating a compliance exposure? Agent Gateway is Tray.ai’s answer to that question — an MCP registry and governance layer that gives IT a single point of visibility and control over every agent-to-system connection in the organization. In a multivendor environment, that visibility is not a convenience. It is a control requirement.
The multivendor reality no one is addressing directly
Here is a finding that should shape every enterprise AI agent evaluation this year: more than half of IT leaders — 51%, per the same Gartner report — are already running a multivendor AI agent strategy. They are not choosing one platform and standardizing on it. They are using multiple models, multiple tools, multiple specialized agents — and then discovering that nothing at the center is holding it together.
Most vendors competing in this market are optimized to win the agent-builder selection. Few of them are built for what comes after: the orchestration and governance layer that makes a multivendor environment manageable rather than chaotic.
Tray.ai’s listing in the Claude Connectors Directory is one concrete example of what genuine interoperability looks like. It means Tray.ai’s 700+ enterprise connectors are available to Claude-based agents — organizations running Anthropic’s models gain access to Tray.ai’s integration infrastructure without changing their AI stack. In an environment where most vendors are optimizing for lock-in, that matters.
For leaders in active evaluation right now
If your organization is in the 42% planning an AI agent deployment this year, the Gartner quadrant is a useful input — but it is not a selection framework. What it tells you is which vendors have the product depth and infrastructure maturity to handle enterprise requirements. What it does not tell you is whether a vendor’s orchestration layer can handle your specific environment: your existing integrations, your governance requirements, your citizen developer population.
The most useful thing you can do before selecting a platform is run a structured evaluation against your actual constraints — not a capabilities demo against a clean dataset, but a live test against the messy, multi-system, permission-layered environment your agents will actually operate in.
See it in action
Run your AI agents on infrastructure built for the enterprise
Talk to us about your orchestration environment — your existing integrations, governance requirements, and citizen developer footprint.
Book a demo →Gartner® Disclaimer
GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally. All rights reserved. Gartner does not endorse any company, vendor, product or service depicted in its publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner publications consist of the opinions of Gartner’s business and technology insights organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this publication, including any warranties of merchantability or fitness for a particular purpose. Emerging Market Quadrant for No-Code Agent Builders — Established Vendors, 8 June 2026, Jason Wong, Keith Guttridge, Eric Goodness, Kelli Smith, Justin Tung