feat: RAG Knowledge Base, Serving Layer, and Model Registry

.env.example

@@ -22,5 +22,36 @@ FORECAST_MAX_HORIZON=90
 FORECAST_MODEL_ARTIFACTS_DIR=./artifacts/models
 FORECAST_ENABLE_LIGHTGBM=false
 
+# RAG Configuration
+# Embedding Provider: "openai" or "ollama"
+RAG_EMBEDDING_PROVIDER=openai
+
+# OpenAI Configuration (when RAG_EMBEDDING_PROVIDER=openai)
+OPENAI_API_KEY=sk-your-openai-api-key-here
+RAG_EMBEDDING_MODEL=text-embedding-3-small
+
+# Ollama Configuration (when RAG_EMBEDDING_PROVIDER=ollama)
+# OLLAMA_BASE_URL=http://localhost:11434
+# OLLAMA_EMBEDDING_MODEL=nomic-embed-text
+
+# Embedding dimension (must match your model: OpenAI=1536, nomic-embed-text=768, etc.)
+RAG_EMBEDDING_DIMENSION=1536
+RAG_EMBEDDING_BATCH_SIZE=100
+
+# Chunking settings
+RAG_CHUNK_SIZE=512
+RAG_CHUNK_OVERLAP=50
+RAG_MIN_CHUNK_SIZE=100
+
+# Retrieval settings
+RAG_TOP_K=5
+RAG_SIMILARITY_THRESHOLD=0.7
+RAG_MAX_CONTEXT_TOKENS=4000
+
+# pgvector index settings
+RAG_INDEX_TYPE=hnsw
+RAG_HNSW_M=16
+RAG_HNSW_EF_CONSTRUCTION=64
+
 # Frontend (Vite)
 VITE_API_BASE_URL=http://localhost:8123