Orbit Launches Agent Builder, a Major Industry Leap Forward in AI Investment Research

Award-winning institutional research platform enables asset managers, hedge funds, and quantitative teams to create self-service agents, each automatically exposed as an MCP endpoint.

LONDON, 03 June 2026 - Orbit Financial Technology, the award-winning AI investment research platform, today announced Agent Builder, a self-service feature within Orbit Insight that enables institutional investment teams to create their own production-grade research agents through conversation. Investment workflows that once required weeks of bespoke development now take minutes. Orbit’s Agent Builder is the first capability of its kind on an institutional research platform. Investment teams own the customization process end-to-end, with no engineering involvement.

Orbit’s Agent Builder addresses one of the most persistent bottlenecks in institutional AI adoption: the gap between buying a research platform and getting it to do exactly what a team needs. Until now, customizing an agent for a specific workflow (a stock screen, a company scorecard, or ongoing sector and competitor analysis) required a consulting engagement, a lengthy specification process, and weeks of iteration. Agent Builder condenses that cycle into a single session, whilst providing the flexibility to tweak and amend to ensure the highest levels of accuracy.

"Investment teams are constantly asking for the flexibility to define their own research workflows without waiting on a vendor. Orbit’s Agent Builder gives them exactly that, built natively into the platform they already use,” said Da Wei, Founder & CEO of Orbit. “The user focuses on the research question; Orbit handles orchestration, data, scheduling, and integration.”

How it works

Every Orbit agent has two components. The first transforms the user's requirement into a configurable, structured data extraction that runs on the back end. The second sits atop that structured data and generates reports. One agent produces one structured dataset and any number of reports from it. Users review the logic at every layer, edit it, and rerun on demand.

Building an agent follows one of two paths. A user can describe the workflow in plain language through a guided conversation. Alternatively, they can upload an existing methodology document in any format, such as a portfolio research process, an ESG policy, a sector dashboard template, a slide deck, or a screenshot. Orbit reads the document, proposes a baseline configuration, and guides the user through review and refinement, layer by layer.

Each agent can pull from Orbit's Knowledge Base datasets, the user's connected internal data, or both. The user selects the combination at build time. Frequency is also user-defined: one-off, scheduled, or always running until stopped.

MCP-native by default

Every agent created in Orbit Insight automatically becomes an MCP endpoint for that user. Teams using Claude, ChatGPT Enterprise, or their own internal copilots can connect directly to a specific agent. The same data flows out via the API in the same self-service way.

This MCP-first design makes Orbit usable as infrastructure within the AI tools investment teams already use, with no additional configuration.

Inside the Agent Marketplace

The Agent Builder lives inside Orbit Insight and the Agent Marketplace, the platform's library of off-the-shelf agents covering earnings analysis, concept screening, peer benchmarking, ESG scoring, thematic analysis, news flow tracker and more. With this launch, the Marketplace gains a new section for each user's personal library of agents they have built themselves. Off-the-shelf and user-built agents are now in one place.

Built on the Orbit Insight platform

The Agent Builder runs on the same foundation as the rest of Orbit Insight. The platform processes 70 million documents per year across 75,000+ companies in 120 countries and 80 exchanges, with exclusive coverage of 5,500 China A share names. Every output is grounded in source documents with paragraph-level citations.

Orbit was selected for the IA Engine Innovator Program 2026 and serves 7 of the top 20 global hedge funds.

About Orbit

Orbit is the award-winning AI investment research platform for institutional buy-side teams. Asset managers, hedge funds, and corporate strategy teams use Orbit to apply their proprietary research frameworks at the scale of every filing, transcript, and news source they need to cover. The platform processes 70 million documents per year across 75,000+ companies in 120 countries and 80 exchanges. Orbit facilitates automated workflows and comprehensive data analysis, streamlining research processes and uncovering critical insights. By committing to data integrity and clarity, Orbit enables users to confidently navigate complex financial landscapes and make strategic decisions with precision. Orbit serves 7 of the top 20 global hedge funds and was selected for the IA Engine Innovator Program 2026.