Skip to Content

Production References

Real systems and frameworks worth studying once the capstone is done.

Real Systems

SystemWhat to look for
Anthropic — Building Effective AgentsOrchestrator–subagent patterns, task decomposition, handoffs
LinkedIn RAG at ScaleHybrid search, retrieval monitoring in production
Uber LLM GatewayModel routing, cost attribution, fallback across providers
Replit AI AgentLong-horizon agent with persistent state and sandboxed tool execution
Lilian Weng — LLM Powered Autonomous AgentsThe most complete survey of agent patterns: planning, memory, tool use
OWASP Top 10 for LLM ApplicationsSecurity risks specific to LLM systems — prompt injection, data leakage, supply chain

Frameworks

Study as reference implementations, not mandatory dependencies.

FrameworkWhat to study
LangGraphStateful graph-based orchestration
LangChainRAG, tool calling, and chain abstractions — large ecosystem, use selectively
LlamaIndexData ingestion, indexing, and retrieval — strongest RAG-specific framework
AutoGenMulti-agent conversations: planner, executor, and critic
CrewAIRole-based agent teams with task delegation
Anthropic Agent PatternsOrchestrator–worker, parallelisation, routing — no framework lock-in
Vercel AI SDKStreaming, tool calling, and UI integration for TypeScript/Next.js
HaystackModular RAG and agent pipelines — good for custom retrieval workflows