AI Core Infra:Semantic Code Search

The Semantic Code Search functional team within AI Core Infra, focused on GitLab’s RAG implementation for providing semantic understanding of code repositories.

Overview

The Semantic Code Search team is part of the AI Core Infra organization. The team focuses on GitLab’s RAG implementation for providing semantic understanding of code repositories.

Key Information

Slack Channel #f_semantic-code-search
Stage Label devops::ai platform
Group Label group::ai core infra
Category Labels category:semantic code search

Team Resources

Resource Link
Team Project gitlab-org/ai-powered/semantic-code-search/team
Planning Issues Planning Issues
Issue Board Issue Board

Runbooks

The Active Context runbook provides operational guidance for the underlying framework powering semantic code search, including the embeddings pipeline, vector store management, and codebase indexing.

Documentation

  1. Semantic Code Search — user documentation
  2. Semantic Search development — development documentation covering the Active Context framework internals

Design Documents

  1. Codebase as Chat Context — feature design (semantic search, embeddings, chunking, ad-hoc indexing)
  2. AI Context Abstraction Layer — ActiveContext framework design (tasks, embedding models)