Recent Blog Posts
Insights on AI Agents, Modern Web, and the Future of Engineering.
Sharing State Across Agent Boundaries: Context Contracts
Define type-safe context contracts with PydanticAI to share validated state safely across multi-agent handoffs and boundary transitions.
Orchestrator-Worker Decomposition: Splitting Complex Tasks Across Agents
Decompose complex multi-step goals into structured worker sub-tasks using the Google Agent Development Kit (ADK) Orchestrator-Worker pattern.
Event-Driven Orchestration: Coordinating Agents over Message Buses
Build asynchronous, event-driven multi-agent systems using message buses like Kafka or RabbitMQ with LlamaIndex Workflows.
Resilience for Orchestrators: Retries, Timeouts, and Circuit Breakers
Build enterprise-grade resilience into multi-agent orchestrators using retries, per-task timeouts, and circuit breakers with the Microsoft Agent Framework.
Fan-Out/Fan-In: Parallel Execution in Multi-Agent Systems
Master the Fan-Out/Fan-In orchestration pattern in LangGraph to run concurrent worker agents in parallel and aggregate multi-agent insights.
Using MCP Tools with the OpenAI Agents SDK
Connect Model Context Protocol (MCP) servers seamlessly into the OpenAI Agents SDK for Python to power autonomous agent workflows with standardized tools.
Testing MCP Servers: Mock Clients and Schema Validation
Build robust unit and integration test suites for Model Context Protocol (MCP) servers using in-memory mock clients, JSON Schema validation, and pytest.
Versioning MCP Servers: Backward-Compatible Tool Interfaces
Maintain backward compatibility and manage API evolution when updating Model Context Protocol (MCP) tool schemas across multi-agent fleets.
Building a Custom MCP Client for Your Agent Runtime
Build a lightweight, production-ready Model Context Protocol (MCP) client from scratch in Python to integrate external tool servers into custom agent frameworks.
Connecting MCP Tools to LangGraph Agents
Integrate Model Context Protocol (MCP) tool servers dynamically into stateful LangGraph agent workflows using langchain-mcp adapters.
Tracing MCP Tool Calls with OpenTelemetry
Instrument Model Context Protocol (MCP) tool executions and agent workflows with OpenTelemetry distributed tracing for end-to-end observability.
Performance Tuning MCP Servers Under Enterprise Load
Optimize Model Context Protocol (MCP) server throughput, lower tool latency, and manage connection pools for high-concurrency enterprise workloads.