AI Workflows

LangGraph 1.x (August 2026) ships node caching, deferred fan-in nodes, pre/post model hooks, and a content-block streaming API. We use all four to cut 68% of redundant compute and 56% of cost on a 40-run/day research pipeline.

Package AI agent capabilities into portable, vendor-neutral Agent Plugins that bundle Skills with MCP servers, then automate schema validation, multi-client smoke tests, and catalog publishing.

Architect a stateless MCP server on Cloudflare Workers that survives restarts between rounds of a human approval gate using MRTR requestState, the tasks extension, and RFC 9207 OAuth verification.

Operate OpenAI's GPT-5.6 programmatic tool calling in production: let Luna, Terra, or Sol author typed JavaScript that orchestrates allowlisted tools inside the OpenAI sandbox, with approval gates, step budgets, and OpenTelemetry tracing.

Implement a robust, real-time AI content moderation pipeline using PydanticAI and Hive Moderation API to secure multi-modal user-generated content.

Deploy a multi-agent AI workflow using CrewAI and Plaid API to automate financial document parsing, deduction classification, and IRS-compliant tax filings.

Unlock the future of machine-to-machine commerce by orchestrating autonomous payment pipelines with Cloudflare Wallets and LangGraph.

Deploy zero-cloud, edge-native medical imaging AI pipelines for rapid and secure diagnostic analysis.

A complete guide to building local, privacy-preserving AI agents using Meta's latest 30B model with LangGraph and Ollama.