AI Workflows & Agentic Architecture Directory

LangGraph Agent Observability Langfuse is a Developer Tools workflow that configures LangGraph v0.2.0 agents to export execution traces, token costs, and state transitions to Langfuse v2.50. The setup uses Python v3.11 a...

LiveKit Gemini voice agent is a production deployment pattern that combines LiveKit Agents SDK v0.10.0 and the Gemini Live API to run low-latency voice calls. This architecture connects WebRTC audio channels directly to...

LiteLLM Proxy Agent Observability is an integration pattern that configures the LiteLLM Proxy v1.60.0 callback system to export model performance and API usage data to Prometheus and Grafana. The system routes requests,...

Deepseek r1 tool calling runs the DeepSeek-R1-Distill-Llama-8B model on a local workstation using Ollama v0.5.0 to execute Python functions without cloud API dependencies. Unlike cloud-based reasoning engines, this archi...

ElevenLabs conversational ai n8n integration is an automation architecture that connects real-time voice agents built with the ElevenLabs Conversational SDK v0.4.0 to enterprise databases and customer relationship manage...

Crewai multi agent hierarchical workflows structure complex process automation by running Python v3.11 tasks under a central manager agent using OpenAI GPT-4o. This orchestration pattern replaces direct agent-to-agent co...

Deepseek r1 local agents run Ollama v0.1.48 and LangChain v0.3.0 on a local workstation to coordinate offline reasoning tasks using open-weight models. Unlike API-dependent setups, this architecture uses Llama.cpp to pro...

A custom MCP server postgres database connection is a local or remote gateway that exposes secure, read-only SQL functions to Model Context Protocol clients like Claude Code. By mapping database schemas to structured too...

Composio tool calling agents integration is a development framework that connects terminal agents like Claude Code v2.1 and OpenAI models to third-party applications via the Composio SDK v0.4.0. The framework provides a...

Crewai market research workflows automate competitive data collection by running Python v3.11 tasks with Firecrawl v1.2.0 and CrewAI Python v0.40.0 under the coordination of OpenAI GPT-4o. This loop collects raw target U...

Claude code vs cursor represents the evaluation of a terminal-based agentic command line interface versus an AI-native integrated development environment editor for software engineering. In developer workflows, this comp...

Build self healing n8n is an automated system that connects n8n v1.52.0 workflows with the Claude Code CLI v0.2.0 terminal agent to detect execution failures, diagnose the root cause, and deploy corrected code blocks in...