Skip to content

Story 3: Asynchronous AI Task Pipeline (Kafka Integration) #3

Description

@KMicha

Story 3: Asynchronous AI Task Pipeline (Kafka Integration)
User Story: As a system architect, I want to establish a Kafka-based "circulatory system" between the backend and AI services so that long-running ML/LLM tasks do not block the web interface.

Task 3.1: Kafka Infrastructure Setup. Update dc-micronaut.yml to include a Kafka broker and Zookeeper node to facilitate asynchronous messaging

.
Task 3.2: Implement LLM Task Producer. Create a Kotlin-based @KafkaClient in Micronaut to publish prompt data and user metadata to an llm-tasks topic
.
Task 3.3: Implement AI Worker (Python). Scaffold a Python service that uses a Kafka poll loop to retrieve messages from llm-tasks, processes them through the local Ollama/Keras routines, and produces the response to an llm-results topic
.
Task 3.4: Implement Backend Result Listener. Create a Kotlin @KafkaListener in the backend to consume messages from llm-results and update the task status in PostgreSQL, triggering a notification to the Vue 3 frontend

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions