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
Configure CrewAI agents to use different LLM providers including Anthropic Claude, local Ollama models, and Azure OpenAI, with model parameter tuning and fallback strategies.
Learn how CrewAI organizes agents into role-based teams with defined tasks, delegation patterns, and process flows to build collaborative AI systems that tackle complex workflows.
Learn how to install CrewAI, define agents with the Agent class, create tasks with the Task class, assemble a Crew, and run it with kickoff to build your first multi-agent workflow.
Configure CrewAI's three memory systems — short-term for session context, long-term for cross-session learning, and entity memory for tracking people and concepts — with storage backends and embedding options.
CrewAI sequential vs hierarchical vs consensual processes: what each one does, when it breaks, and how to pick the right process for your crew.
Deploy CrewAI crews to production with Docker containerization, implement robust error handling with retry strategies, track costs, and optimize performance for scalable multi-agent systems.
Master CrewAI Task design including task structure, expected_output specifications, context chaining between tasks, and async task execution for parallel agent workflows.
Extend CrewAI agents with built-in tools like SerperDevTool and ScrapeWebsiteTool, create custom tools using the @tool decorator, and configure tool sharing across multiple agents.
Build an AI agent that synchronizes data between CRMs like Salesforce and HubSpot, automatically logs activities, updates pipeline stages, and enriches contact records using intelligent data processing.
