Skip to content

Releases: maxzaikin/TGB-MicroSuite

Release v0.2.0

29 Jun 12:17
14fdb00

Choose a tag to compare

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog,
and this project adheres to Semantic Versioning.

[0.2.0] - 2025-06-29

Added

  • Introduced MLOps Orchestration with ZenML: Integrated a self-hosted ZenML server into the core infrastructure (docker-compose.infra.yml) to manage and track ML workflows.
  • New Reproducible Data Ingestion Pipeline: Implemented the feature_ingestion_pipeline, a formal, multi-step ZenML pipeline to replace the previous manual script. This provides full reproducibility and observability for our RAG data processing.
  • Persistent MLOps Metadata: Added a dedicated Docker volume (zenml-db) to ensure all pipeline run history, artifacts, and metadata are persisted across container restarts.
  • Detailed MLOps Documentation: Created a new MLOPS_README.md file with a complete guide on the pipeline architecture and usage instructions.

Changed

  • Data Ingestion Workflow: The process for ingesting documents into the RAG knowledge base has been fundamentally changed. It is now exclusively managed by the new ZenML pipeline, ensuring all data operations are versioned and tracked.

Fixed

  • Documentation Rendering: Corrected Mermaid diagram syntax in MLOPS_README.md to prevent Unsupported markdown warnings on GitHub.

Removed

  • Legacy Ingestion Script: Deleted the old scripts/ingest.py script and its associated README-ingest.md to eliminate redundancy and create a single source of truth for data ingestion.