Skip to content

Talk Proposal: Build production-ready ambient agents using Drasi and Dapr Agents #28

Description

@JeffreyJPZ
  • Draft Title: Build production-ready ambient agents using Drasi and Dapr Agents
  • Length:
    • Lighting talk/Demo (10 min)
    • Full Presentation (20 min)
  • Links: (slides, article, notes)
  • Which dates, particularly Tue/Wed/Thu during the 3rd/4th week of a month, are you likely available? (e.g. "Most Thursdays with 4 weeks notice", "June or later", etc.)
  • Short Summary of your talk:
    • While AI agents may work well enough for local demos, moving to long-running autonomous agents that interact with other services (internal microservices, third party APIs, etc.) raises problems of scalability, security, durability, resiliency, auditability, and governance, especially in highly-regulated industries. Triggering these workflows when complex business conditions are met is another problem — internal services and third-party providers not be equipped to provide the context necessary for agents, and data is often spread across multiple sources which makes custom change processing pipelines challenging to maintain.

    • In this talk, I'll give a brief overview of the "high-level" change detection capabilities of Drasi and the microservices APIs provided by Dapr, how they address the mentioned problems, and how they can be combined to help developers and operators build change-driven, "ambient" agents declaratively. I'll walk through a live demo showing how Dapr Agents can be triggered from Drasi "business events" in a Kubernetes environment, thanks to Dapr Agents' inherent support for asynchronous messaging and event-driven workflows. Lastly, I'll give a sneak peek at the upcoming work for the Google Summer of Code project as the focus shifts from purely reactive agents to autonomous, proactive agents.

Speaker Bio

Metadata

Metadata

Labels

Projects

No projects

Relationships

None yet

Development

No branches or pull requests

Issue actions