Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Build a production-ready end-to-end guardrail pipeline combining input validation, output sanitization, and tool-level guardrails using the OpenAI Agents SDK with full working code.
Master the @function_tool decorator in the OpenAI Agents SDK. Learn how to create sync and async tools, handle complex parameters, and wire multiple custom tools into your agents.
Learn how to build custom guardrails using agent-based validation, confidence thresholds, and multi-layer tripwire strategies in the OpenAI Agents SDK for production-grade safety.
Build a production-ready MCP server that wraps your existing REST API endpoints as callable tools, using FastAPI and the MCP Python SDK to expose your business logic to AI agents.
Build custom trace processors and exporters for the OpenAI Agents SDK to ship agent telemetry to Elasticsearch, Datadog, or any backend using TraceProvider, BatchTraceProcessor, and BackendSpanExporter.
A complete end-to-end tutorial building a production-grade customer support system with triage, billing, refund, and FAQ agents using handoffs, RECOMMENDED_PROMPT_PREFIX, and prompt_with_handoff_instructions in the OpenAI Agents SDK.
Beginner-friendly walkthrough of building a complete agentic AI app — from project setup and agent creation to testing and deployment. Progressive complexity.
Step-by-step tutorial to build a working voice agent using OpenAI's VoicePipeline — from installing dependencies and capturing microphone audio to streaming agent responses through your speakers.
Build a knowledge base chat agent using OpenAI's vector stores API for document upload, chunking, semantic search, citation-grounded answers, and automatic knowledge base maintenance.
