The bridge for agentic banking

MidLyr is building the infrastructure necessary to make traditional banking systems compatible with the era of AI agents. The central premise of the company is that consumers are increasingly using general-purpose AI assistants to reason about their lives, including their finances, but banks are currently locked out of these conversations. By the time a customer logs into a mobile banking app, the decision-making process—mediated by an LLM—has often already finished. MidLyr provides the connective tissue that allows banks to plug their existing APIs directly into these AI assistants safely and at scale.

The core product is a platform that manages the "mid-layer gap." In a typical banking environment, every new product or channel requires a complex web of integrations for fraud, compliance, KYC, and disputes. MidLyr consolidates these operations into a single platform that interfaces with external AI models. Technically, this is achieved by generating a Model Context Protocol (MCP) server that AI assistants can call. This approach allows an LLM to interact with a bank's internal systems using a standardized framework, reducing the integration time from the typical 12-18 months of in-house development to a window of two to six weeks.

Solving the authentication bottleneck

The most significant hurdle in AI-mediated banking is not the data retrieval itself, but the verification of intent. MidLyr focuses heavily on "Bank-Grade AI Authentication," which addresses the problem of ensuring that an AI agent is truly authorized to act on behalf of a human customer. Their system uses customer-approved action tokens and encrypted session management. For high-risk operations, such as moving money or changing account settings, the platform triggers out-of-band confirmations via SMS or email, ensuring that the human remains the final authority in an otherwise automated workflow.

From a regulatory perspective, MidLyr is designed to satisfy examiners from bodies like the OCC, CFPB, and FCA. The platform generates a complete audit trail for every AI-mediated request, logging the prompt, the agent's reasoning, and the final system action. This level of transparency is intended to move AI banking from a experimental "lab" phase into production environments where compliance is a non-negotiable requirement. The software can be deployed in a Virtual Private Cloud (VPC) or on-premises, catering to the strict data residency requirements of global financial institutions.

Market position and origins

Founded in 2025 by Wael Elsahhar and Ruochen Ren, MidLyr recently emerged with a $2.5 million pre-seed round led by investors including Story Ventures and Hustle Fund. The team is Egyptian-led and has a strong focus on the MENA region, though the product is designed for the global financial services market. They are entering a space where the primary competition is not other startups, but the status quo of internal bank IT departments. By positioning themselves as an "off-the-shelf" solution for AI compliance and connectivity, they are betting that banks will prefer a specialized vendor over the risk of building custom AI gateways from scratch. The company is SOC 2 Type II certified and maintains high uptime SLAs, signaling its intent to be viewed as enterprise infrastructure rather than a peripheral tool.