Upgrade to Pro — share decks privately, control downloads, hide ads and more …

apidays New York 2025 - Building Agentic Workf...

apidays New York 2025 - Building Agentic Workflows with FDC3 Intents by Nick Kolba (Connectifi)

Building Agentic Workflows with FDC3 Intents
Nick Kolba, Co-founder & CEO at Connectifi

apidays New York 2025
API Management for Surfing the Next Innovation Waves: GenAI and Open Banking
May 14 & 15, 2025

------

Check out our conferences at https://www.apidays.global/

Do you want to sponsor or talk at one of our conferences?
https://apidays.typeform.com/to/ILJeAaV8

Learn more on APIscene, the global media made by the community for the community:
https://www.apiscene.io

Explore the API ecosystem with the API Landscape:
https://apilandscape.apiscene.io/

Avatar for apidays

apidays

May 24, 2025
Tweet

More Decks by apidays

Other Decks in Programming

Transcript

  1. About Nick Kolba Co-founder and CEO of Connectifi. Prior to

    Connectifi, Nick held a number of leadership positions in financial technology companies both large and small, including CTO at OpenFin where he led the creation of the FINOS FDC3 standard. Connectifi We connect all of the tools for an enterprise so that they can seamlessly interoperate and share functions and data. Connectifi empowers end users with the flexibility to choose the best path for their workflows in real-time and organizations with the ability to move fast and scale securely and efficiently.
  2. Inspired by Mobile Intents Loosely Coupled Integration Across Disparate Applications

    Dynamic Discovery of Next Best Actions FDC3 Intent–based interoperability Workflows determined at runtime, based on user goals and context. Orchestration of User-Centric Workflows
  3. Historically, connecting an AI model to all these external sources

    has been a messy, ad-hoc affair. Developers had to write custom code or use specialized plugins for each data source or API. This made “wire together” integrations brittle and hard to scale. “ Ksenia Se, HuggingFace ” It’s clear that there needs to be a standard interface for execution, data fetching, and tool calling. APIs were the internet’s first great unifier—creating a shared language for software to communicate — but AI models lack an equivalent. “ “” Yoko Li, Andreessen Horowitz With "the prompt is the program," we can build more adaptable, intention-based integrations that focus on what needs to be done rather than how it's technically implemented. “ “” Steve Manuel, Dylibso Common Ground: FDC3 & Agentic Emerging standards such as MCP & A2A have made it clear - interoperability is key to agentic AI
  4. Unlike humans, LLMs are really good at understanding the structured

    data of FDC3. FDC3 Application and AI Agent parity FDC3 Intents and Context inferred from unstructured prompts LLM reasoning across intents to produce workflow The FDC3 to Agentic Journey
  5. Message Flow - FDC3 App App Interop Service Raise Intent

    Intent & Context matching Return Ambiguous Results User Chooses Option Register Intent Intent Routed Intent Result Intent Resolved
  6. Message Flow - Agentic FDC3 Chat App Interop Service User

    Prompt User Chooses Option Register Intent Intent Routed Intent Result Intent Resolved Agent Infer Intent & Context Raise Intent Intent & Context matching Return Ambiguous Results Context Mapping Handle Result
  7. Putting It All Together Dynamic Tool Use, Multi-Agent Workflows, HiL

    Automation Interoperability Service End user FDC3 Directory Agent LLMs Tools FDC3 SaaS 8 Other Agents APPS MCP Other Tools
  8. Interoperability Lessons Learned Demos are easy, meaningful implementation is hard

    Line-of-business sponsorship is essential for adoption Scaling with security, observability, and repeatability is crucial