Forge Documentation
The Autonomous Data Infrastructure Platform
Forge AI Intelligence Stack
Forge is powered by a 4-layer AI architecture that provides schema intelligence, governance automation, schema evolution, and autonomous orchestration.
Merlin — Autonomous Data Agent
Flagship • Q4 2026
"Set up my data pipelines and maintain them."
Merlin is an LLM-powered autonomous agent that understands natural language goals, plans
multi-step workflows, orchestrates Forge's AI tools, and self-heals failures without
human intervention.
Excalibur
Production
Schema Classification
Graph Neural Network (GNN) that treats schemas as graphs to classify data
patterns. Privacy-preserving design—field names never leave your environment.
- GraphSAGE GNN + RandomForest
- 89% accuracy, 5 categories
- Privacy-first fingerprinting
Llamrei
Q2 2026
Schema Evolution
Automatically detects legacy API versions and normalizes them to modern golden
schemas. Saves $200K-$500K per avoided migration.
- 50+ API golden schemas
- Stripe, Salesforce, Shopify
- Non-destructive transforms
Pridwen
Production
Governance ML
Hybrid 3-layer system (Rules + ML + Crowd) that automatically detects PII and
recommends transformations like hash, mask, and encrypt.
- 15 SQL templates
- Day-1 intelligence
- Network effects learning
Core Platform Features
Multi-Warehouse
One parse generates dbt models for BigQuery, Snowflake, Databricks, and Redshift.
Deep JSON Unnesting
Automatically flattens nested arrays and objects 5+ levels deep into relational tables.
dbt Model Generation
Production-ready SQL models with proper keys, types, and documentation.
2026 Roadmap
Forge Core
4-warehouse support, Excalibur GNN, Pridwen governanceLlamrei
Schema evolution, golden schema libraryMerlin Beta
Autonomous agent, natural language goalsMerlin GA
Full autonomy, self-healing pipelinesQuick Actions
Need Help Getting Started?
Our team is here to help you transform your JSON data into production-ready, AI-governed analytics.
View Tutorial Get Support