"Partnering with Microsoft’s AI Co-Innovation Lab provided critical technical resources and expertise that accelerated our project timelines and validated our approach to AI in commercial finance.” – Riccardo Pietri, VP of Engineering & Security, Trade Ledger
Founded in 2016, Trade Ledger provides clients with a working capital platform that revolutionizes business lending, powered by the latest smart technologies and digital datasets. Trader Ledger uses cloud-native, API-first technology that supports customer onboarding, credit origination, risk assessment, and portfolio management, enabling lenders to quickly launch and manage new products.
As an industry disrupter, Trade Ledger has capitalized on opportunities in AI technologies, responding to the need for faster decision-making tools to combat historical industry challenges, like manual data handling, fragmented systems, and a limited capacity for real-time analysis.
Most recently, they saw an opportunity to leverage large language models (LLMs) and build an AI copilot that would provide intelligent insights to customers, significantly reducing time to decision.
They joined forces with the AI Co-Innovation Lab in Munich to help them tackle the technical complexities of such a solution, including:
The AI Co-Innovation Lab team and Trade Ledger came together to develop a new version of a working capital copilot, ensuring the solution would be more effective, intuitive, and interactive. The Lab team also shared essential AI knowledge to assist the Trade Ledger team in choosing an approach to deploy.
The teams came up with three different state-of-the-art AI frameworks for a copilot that could use natural language to speak to various databases and pull the appropriate data to build customer reports and insights. The Lab team leveraged Azure OpenAI Service to build a model that could streamline the data connected to large language models (LLMs) and generate timely and accurate information.
They also ensured the copilot could be distributed via collaboration services like Microsoft Teams and Microsoft 365 Copilot, where users could both interact intuitively with the AI and collaborate with stakeholders and colleagues.
Another essential part of Trade Ledger’s experience was the learning sessions with the Lab team around cloud engineering, cloud architectures using gen AI, and various gen AI platforms, designed to empower Trade Ledger to deploy their chosen solution after their engagement in the Lab.
By integrating with Microsoft’s advanced AI models and infrastructure, this copilot enhancement can automate data retrieval, process natural language queries, and interpret financial data more effectively. These capabilities are particularly useful for financial services clients who need quick, accurate, and structured insights.
The Munich Lab accelerated Trade Ledger’s time to deployment with three approaches to their copilot. This will enable Trade Ledger to implement their AI copilot with a major global bank as a beta client in 2025, where they can validate the solution’s real-world effectiveness. Trade Ledger’s future goal is to develop more robust use cases that will help them build an AI portfolio that redefines end-to-end trade and working capital finance for banks globally.