Skip to content

Fluxnova General Meeting - 20 JAN 2025 #57

@TheJuanAndOnly99

Description

@TheJuanAndOnly99

Date

Tuesday JAN 20 2025 - 10am (US EST) / 3pm (London, GMT)

Zoom info

Join Zoom Meeting:
https://zoom-lfx.platform.linuxfoundation.org/meeting/98975602417?password=8a1c796a-2eba-4f71-beeb-e3ecb7eb3450

Register to the meeting series:
https://zoom-lfx.platform.linuxfoundation.org/meeting/98975602417?password=8a1c796a-2eba-4f71-beeb-e3ecb7eb3450&invite=true

Meeting notices

  • FINOS Project leads are responsible for observing the FINOS guidelines for running project meetings. Project maintainers can find additional resources in the FINOS Maintainers Cheatsheet.

  • All participants in FINOS project meetings are subject to the LF Antitrust Policy, the FINOS Community Code of Conduct and all other FINOS policies.

  • FINOS meetings involve participation by industry competitors, and it is the intention of FINOS and the Linux Foundation to conduct all of its activities in accordance with applicable antitrust and competition laws. It is therefore extremely important that attendees adhere to meeting agendas, and be aware of, and not participate in, any activities that are prohibited under applicable US state, federal or foreign antitrust and competition laws. Please contact [email protected] with any questions.

  • FINOS project meetings may be recorded for use solely by the FINOS team for administration purposes. In very limited instances, and with explicit approval, recordings may be made more widely available.

Agenda

  1. External AI systems calling Fluxnova workflows as tools
  • The idea is to allow external AI agents to invoke Fluxnova workflows directly
  • For example: a process uses an MCP style start event and once deployed, is exposed as an MCP tool endpoint that any internal AI agent can call.

(This was presented by Pieter. If the community is aligned, NatWest would like to take this away, package it up and contribute back.)

  1. Fluxnova as the place where Agentic AI happens
  • Similar to the C8 approach
  • AI steps are modeled as sub-processes
  • The AI determines the next path, with huma in the loop option
  • ...
  • AOB & Adjourn (5mins)

Minutes

Note that the minutes were generated by AI and may not be fully accurate

Quick Recap

The meeting began with introductions and administrative updates, including a reminder about FINOS Meeting Notices and guidelines. The team discussed upcoming events, including a Fluxnova webinar and a hackathon, while reviewing the development roadmap and exploring AI integration opportunities. The session concluded with demonstrations of AI capabilities and discussions about community involvement in project development, with emphasis on supporting new participants and reviewing roadmap priorities.


Next Steps

  • Olivier: Share the template for the Fluxnova webinar with the group.
  • Chris and Ryan: Participate in the Fluxnova webinar practice session by end of this week.
  • Edson: Sync up with the tech lead regarding the monitoring UI decoupling progress and keep the group posted on status.
  • Edson: Create and present a simple PowerPoint or overview highlighting areas of the monitoring UI where community input or development could be valuable; Rachana Ramshetty to remind Edson.
  • Olivier: Coordinate with Pieter (and potentially Aurelian or Peter) to agree on what to present at the MCP Dev Summit and prepare the submission before the call for papers closes.
  • Pieter: Come back to Olivier to discuss and potentially collaborate on MCP Dev Summit presentation.
  • Juan: Fix the misspelling of "Fluxnova" in the title of GitHub Discussion 59.
  • Chris: Demo the AI agent workflow implementation at the next (Thursday/Friday) call.
  • Chris and Ryan: Meet offline to discuss the use of CMMN for ad hoc subprocess capabilities in Fluxnova.
  • All interested contributors: Review the AI feature discussion thread (link in chat) and add ideas, upvote, or ask questions to help prioritize next steps.
  • Chris and Ryan: Confirm whether CMMN module is supported in current Fluxnova and report findings to the group.
  • Edson: Keep the group updated on the progress of decoupling the monitoring UI and notify when the code is ready for community review.

Summary

Fluxnova Project Team Introductions

The meeting began with introductions from various participants, including new members from several organizations interested in the Fluxnova project. It was explained that the meeting was a merged session of the general call and the AI working group call to optimize time. Attendees were reminded of the FINOS Meeting Notices, including guidelines on antitrust policies and community conduct. The session concluded with an open invitation for further introductions before transitioning into the main agenda.

Fluxnova Webinar and Hackathon Updates

The meeting covered two main announcements: a Fluxnova webinar scheduled for January 28th, with organizers planning a practice session by week's end, and a DTCC hackathon running February 23-26. The hackathon was noted as particularly important due to available AI-focused grants. Participants were informed they would receive email communication regarding the webinar, and a planned AI working group demo was showcased during the final 30 minutes of the meeting.

Fluxnova Development and Summit Plans

The team discussed two main topics: a demo competition with a grant opportunity and the upcoming MCP Dev Summit. Offers were made to present at the summit, and interest was expressed regarding collaboration on the presentation. The group also reviewed the Fluxnova development roadmap; it was explained that the monitoring UI would be decoupled from organization-specific aspects by the end of Q1. Agreement was reached to explore opportunities for community involvement in developing isolated parts of the project. Finally, two priorities for AI integration in Fluxnova were touched upon, as previously discussed in the AI working group context.

AI Integration in Fluxnova

The team discussed two main workstreams for AI integration in Fluxnova: improving developer experience through generative AI tools and using Fluxnova to orchestrate AI agents. A proof-of-concept implementation using BPMN subprocesses and Java delegates was demonstrated and will be shown in an upcoming webinar. The team explored the possibility of using CMMN capabilities to support ad-hoc processes, with suggestions made that this could be a viable alternative. Emphasis was placed on Fluxnova being community-maintained, and support information was provided for new participants. The conversation ended with a reminder for participants to review the roadmap and contribute to ongoing discussions about AI features.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    Status

    Backlog

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions