Haggai Roitman published a new book on June 22, 2026, titled The Hitchhiker's Guide to Agentic AI, serving as a practitioner's reference for building autonomous AI systems. The work provides a full-stack technical guide, arguing that developing high-quality agentic systems requires mastery across every layer of the development pipeline rather than focusing on a single component. The first half establishes core foundations including transformer architectures, GPU utilization, model compression, and inference optimization, alongside alignment methods such as RLHF (reinforcement learning from human feedback), PPO, and DPO.

The second half shifts focus to agentic AI, covering specific modules such as retrieval-augmented generation (RAG), memory systems, and multi-agent coordination architectures. The book details protocols for inter-agent communication, including the Model Context Protocol (MCP) and Agent-to-Agent (A2A) standards. It also offers guidance on agent design patterns, evaluation methodologies, and production deployment. Every chapter integrates theoretical concepts with implementation code examples and references to primary literature to support developers throughout the agent-building lifecycle.

Haggai Roitman published a new book on June 22, 2026, titled The Hitchhiker's Guide to Agentic AI, serving as a practitioner's reference for building autonomous AI systems. The work provides a full-stack technical guide, arguing that developing high-quality agentic systems requires mastery across every layer of the development pipeline rather than focusing on a single component. The first half establishes core foundations including transformer architectures, GPU utilization, model compression, and inference optimization, alongside alignment methods such as RLHF (reinforcement learning from human feedback), PPO, and DPO.

The second half shifts focus to agentic AI, covering specific modules such as retrieval-augmented generation (RAG), memory systems, and multi-agent coordination architectures. The book details protocols for inter-agent communication, including the Model Context Protocol (MCP) and Agent-to-Agent (A2A) standards. It also offers guidance on agent design patterns, evaluation methodologies, and production deployment. Every chapter integrates theoretical concepts with implementation code examples and references to primary literature to support developers throughout the agent-building lifecycle.