Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Opey: Agentic Payments via Consent
See a live demo of Opey, an AI agent for banking APIs, executing financial tasks like payments and account creation using OpenAPI and consent.
Opey is an AI agent built for banking APIs that can move from “guidance” to “execution”. With delegated permissions via granular consents, it can directly execute API calls in regulated environments.
In this talk we’ll live-demo Opey orchestrating practical financial-service tasks — initiating a payment, creating customer accounts, and retrieving customer information — while showing the underlying pattern that makes this feasible in regulated environments: OpenAPI/Swagger-driven endpoint discovery + tool calling, under consent/auth constraints.
We’ll show the endpoints Opey selects, the calls it executes, and how consent/auth gates the workflow, so the room can discuss real-world trade-offs and not just hypothetical agent design.
LangGraph-powered agentic chatbot integrates Open Bank Project API with retrieval evaluation.
Open source API platform manages secure, compliant transactions adhering to global Open Banking standards.
- Open Bank Project APIThe Open Bank Project (OBP) is the leading open-source API platform: it delivers a RESTful interface for banks to securely launch financial-grade APIs and power a global FinTech ecosystem.The Open Bank Project (OBP) is the global standard and API solution for the financial services industry, pioneering open banking concepts since 2010. It provides a RESTful interface, allowing developers to build customer-facing applications—like PFM solutions and online accounting integrations—without coding for disparate bank systems. OBP enables financial institutions to create, test, and manage APIs that ensure compliance with major standards (e.g., UK Open Banking, Berlin Group). The platform unlocks a wide range of use cases: this includes secure access to account information, initiating payments, and enriching transactions with metadata (tags, geolocation). OBP has advised over 100 banks and regulators globally, proving its robust, production-ready capability.
- OBP Opey IIOBP Opey II is the Open Bank Project's AI-powered API agent, leveraging GPT-4o to streamline API discovery, troubleshooting, and execution for banks and developers via natural language.Opey II (Open Bank Project Opey II) is an intelligent, model-agnostic API assistant: it transforms complex banking API interactions into simple, natural language conversations. Built by TESOBE, Opey is modeled on advanced Large Language Models (like GPT-4o) and acts as a plugin to the OBP API Explorer II, providing instant guidance on endpoint usage, authentication, and technical operations. This agent accelerates API discovery and integration, offering a non-user interface that dynamically synthesizes data and suggests API call combinations to satisfy specific business or technical use cases for both newcomers and seasoned developers.
- LangraphLangGraph is the low-level, open-source orchestration framework from LangChain: it builds, manages, and deploys complex, stateful AI agent workflows using a graph-based architecture.LangGraph, an open-source framework from the LangChain team, delivers the robust, low-level orchestration required for production-ready AI agents. It leverages a graph-based architecture (nodes and edges) to model and manage intricate, stateful workflows, ensuring reliable, multi-step reasoning. This design supports critical features: durable execution, comprehensive memory, and human-in-the-loop oversight. Companies like Klarna, Replit, and Elastic rely on LangGraph to deploy sophisticated agent systems, such as multi-agent chatbots or complex decision-making engines, with complete control over the cognitive architecture.
- OpenAPIOpenAPI Specification (OAS) provides a vendor-neutral, language-agnostic interface description format for modern REST APIs, using JSON or YAML.OpenAPI defines a standard, machine-readable format for describing HTTP APIs: this is the core value proposition. The specification, currently at version 3.1.0, allows both humans and computers to understand a service's capabilities without source code access. An OpenAPI Description (OAD) drives critical tooling: it enables automated documentation generation, facilitates client and server code creation (SDKs), and streamlines testing workflows. Governed by the OpenAPI Initiative (OAI) under the Linux Foundation, it ensures a robust, open standard, maximizing interoperability and minimizing guesswork in API consumption.
- SwaggerSwagger is the definitive open-source toolset for designing, documenting, and consuming REST APIs (OpenAPI Specification).Swagger is a robust collection of open-source tools centered on the OpenAPI Specification (OAS), formerly the Swagger Specification. It streamlines the API lifecycle: use the Swagger Editor to define your API in YAML or JSON; leverage Swagger UI to generate interactive, web-based documentation for testing endpoints; and utilize Swagger Codegen to automatically generate client libraries and server stubs in various languages (e.g., Java, Python). This toolchain ensures precision, accelerates development, and drastically improves API discoverability for both internal teams and external consumers.
Related projects
Agent discovery and signals in OBP
Milan
This talk explores agent discovery and signals within OBP, offering insights into how agents find and communicate information.
Separating Intelligence from Authority: Deterministic Control Patterns for Multi-Agent Systems in Regulated Finance
Milan
Learn how to build production multi-agent systems for regulated finance, separating AI intelligence from deterministic authority using structured…
From paper to strategy - a Nerovynn experiment
Milan
Discover how to rapidly test financial strategies and manage portfolios using an agent that downloads, extracts knowledge from,…
Engineering Guardrailed AI for Constrained Environments
Milan
Learn how to implement guardrailed autonomy for AI agents, ensuring they operate safely within company policies and security…
Operationalising FINOS AI Governance Framework
Milan
Learn how the FINOS AI Governance Framework bridges policy and production, integrating with emerging standards and orchestration engines…
Continuously Learning AI in Regulated Environments: Versioned Models, Secure Execution, and Controlled Adaptation
Milan
Learn how to build continuously learning AI in regulated environments. This talk details versioned models, secure execution, and…