From quickstart
to production.
Deep-dive tutorials and architecture guides that take you from first API call to production-grade AI applications.
Building a Production RAG Pipeline
A complete walkthrough of a production-grade retrieval-augmented generation pipeline — from document ingestion through semantic chunking, hybrid search, re-ranking, and streaming responses to your application.
Getting Started
Your First Agent in 10 Minutes
Install the SDK, authenticate, and run your first agent workflow end-to-end.
Understanding Domain Packs
What Domain Packs are, how they're structured, and when to use them vs. custom agents.
Authentication & API Keys
Managing API keys, rotating credentials, and setting up environment-specific deployments.
Agents
Multi-Agent Workflow Patterns
Supervisor-worker patterns, parallel fan-out, critic loops, and when to use each.
Tool Use & Function Calling
Defining tools, handling tool outputs, and managing tool call chains safely.
Error Handling & Fallbacks
Retry policies, fallback paths, and building agents that recover gracefully from failures.
Streaming Agent Outputs
Streaming partial outputs, handling stream events, and building responsive UIs on top of agent streams.
Retrieval & Memory
Building a Production RAG Pipeline
Complete guide to semantic chunking, hybrid search, metadata filtering, and re-ranking.
Choosing a Memory Strategy
When to use short-term buffers vs. episodic memory vs. semantic retrieval — and how to combine them.
Connecting Your Vector Store
BYOVS guide: Pinecone, Weaviate, Qdrant, and pgvector adapter configuration.
Document Ingestion at Scale
Batch ingestion pipelines, incremental updates, versioning your knowledge base, and handling large file types.
Guardrails & Compliance
Configuring PII Redaction
Field-level redaction, masking strategies, and HIPAA/GDPR compliance patterns.
Building a Role-Based Policy
Define output policies that change based on authenticated user role and context.
Audit Logging & Export
Configure tamper-evident audit logs, set retention policies, and export to S3 or your SIEM.
Production & Deployment
Deployment Environments & Versioning
Managing dev/staging/prod environments, pack versioning, and zero-downtime upgrades.
Observability & Monitoring
Trace events, token usage monitoring, latency dashboards, and alerting on agent failures.
Rate Limits & Scaling
Understanding rate limits, scaling patterns, and request queuing for high-throughput workloads.