https://hhy.homes All posts from Craig — tech articles and essays. en Wed, 16 Sep 2026 00:31:00 GMT https://hhy.homes/en/posts/hitchhiker-agentic-ai-30-conclusion-and-future-directions https://hhy.homes/en/posts/hitchhiker-agentic-ai-30-conclusion-and-future-directions Wed, 16 Sep 2026 00:31:00 GMT This guide has traced the full arc from transformer foundations through reinforcement learning for alignment to the construction of autonomo… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-29-quick-reference https://hhy.homes/en/posts/hitchhiker-agentic-ai-29-quick-reference Wed, 16 Sep 2026 00:30:00 GMT This chapter consolidates key equations, architecture specifications, API references, and failure mode diagnostics for rapid lookup during d… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-28-quiz-questions-detailed-answers https://hhy.homes/en/posts/hitchhiker-agentic-ai-28-quiz-questions-detailed-answers Wed, 16 Sep 2026 00:29:00 GMT This chapter provides a comprehensive set of questions designed to test and reinforce your understanding of the material covered throughout… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-27-agentic-ui-frameworks https://hhy.homes/en/posts/hitchhiker-agentic-ai-27-agentic-ui-frameworks Wed, 16 Sep 2026 00:28:00 GMT As large language models transition from passive text generators to active agents capable of planning, tool use, and multi-step reasoning, t… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-26-agent-development-frameworks https://hhy.homes/en/posts/hitchhiker-agentic-ai-26-agent-development-frameworks Wed, 16 Sep 2026 00:27:00 GMT The transition from a research prototype to a production-grade agent system is one of the most demanding engineering challenges in modern AI… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-25-multi-agent-systems https://hhy.homes/en/posts/hitchhiker-agentic-ai-25-multi-agent-systems Wed, 16 Sep 2026 00:26:00 GMT The history of artificial intelligence is, in many ways, a history of scale.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-24-agent-to-agent-communication-a2a https://hhy.homes/en/posts/hitchhiker-agentic-ai-24-agent-to-agent-communication-a2a Wed, 16 Sep 2026 00:25:00 GMT As large language models evolve from isolated assistants into collaborative networks of specialized agents, the question of how agents talk… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-23-agent-skills https://hhy.homes/en/posts/hitchhiker-agentic-ai-23-agent-skills Wed, 16 Sep 2026 00:24:00 GMT As agents evolve from monolithic prompt-and-tool systems into modular architectures, a key design question emerges: how should an agent’s ca… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-22-model-context-protocol-mcp https://hhy.homes/en/posts/hitchhiker-agentic-ai-22-model-context-protocol-mcp Wed, 16 Sep 2026 00:23:00 GMT The rise of tool-augmented language models has created a fragmentation problem: every agent framework, every LLM provider, and every enterpr… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-21-agentic-environments-and-benchmarks https://hhy.homes/en/posts/hitchhiker-agentic-ai-21-agentic-environments-and-benchmarks Wed, 16 Sep 2026 00:22:00 GMT The evaluation of a conversational language model is, in principle, straightforward: present a prompt, collect a response, and score it agai… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-20-agent-design-patterns https://hhy.homes/en/posts/hitchhiker-agentic-ai-20-agent-design-patterns Wed, 16 Sep 2026 00:21:00 GMT Building effective agents requires more than a powerful model and a set of tools.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-19-loop-engineering https://hhy.homes/en/posts/hitchhiker-agentic-ai-19-loop-engineering Wed, 16 Sep 2026 00:20:00 GMT The evolution of how practitioners interact with LLM-based agents has followed a remarkably consistent trajectory.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-18-agent-harness-context-management-and-orchestrati https://hhy.homes/en/posts/hitchhiker-agentic-ai-18-agent-harness-context-management-and-orchestrati Wed, 16 Sep 2026 00:19:00 GMT Modern LLM-based agents do not operate in isolation.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-17-agentic-memory-systems https://hhy.homes/en/posts/hitchhiker-agentic-ai-17-agentic-memory-systems Wed, 16 Sep 2026 00:18:00 GMT Large language models are, at their core, stateless function approximators: given a prompt x, they produce a distribution over continuations… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-16-retrieval-augmented-generation-rag https://hhy.homes/en/posts/hitchhiker-agentic-ai-16-retrieval-augmented-generation-rag Wed, 16 Sep 2026 00:17:00 GMT Retrieval-Augmented Generation (RAG) [209] has emerged as one of the most practically impactful techniques for deploying large language mode… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-15-introduction-to-agentic-ai https://hhy.homes/en/posts/hitchhiker-agentic-ai-15-introduction-to-agentic-ai Wed, 16 Sep 2026 00:16:00 GMT The previous parts equipped us with the algorithmic toolkit—how to train, align, and reason with LLMs.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-14-llm-evaluation https://hhy.homes/en/posts/hitchhiker-agentic-ai-14-llm-evaluation Wed, 16 Sep 2026 00:15:00 GMT Evaluation is the backbone of any rigorous machine learning pipeline, yet it is perhaps the most underappreciated component in the developme… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-13-rl-for-large-reasoning-models https://hhy.homes/en/posts/hitchhiker-agentic-ai-13-rl-for-large-reasoning-models Wed, 16 Sep 2026 00:14:00 GMT The emergence of large reasoning models represents one of the most significant developments in modern AI.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-12-llm-agentic-training https://hhy.homes/en/posts/hitchhiker-agentic-ai-12-llm-agentic-training Wed, 16 Sep 2026 00:13:00 GMT Modern LLMs are increasingly deployed not just as conversational assistants but as autonomous agents that interact with external tools, APIs… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-11-system-architecture-infrastructure-at-scale https://hhy.homes/en/posts/hitchhiker-agentic-ai-11-system-architecture-infrastructure-at-scale Wed, 16 Sep 2026 00:12:00 GMT Training LLMs with reinforcement learning from human feedback is as much a systems engineering challenge as it is an algorithmic one.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-10-sft-best-practices-and-techniques https://hhy.homes/en/posts/hitchhiker-agentic-ai-10-sft-best-practices-and-techniques Wed, 16 Sep 2026 00:11:00 GMT Supervised Fine-Tuning (SFT) is the foundation of the RLHF pipeline.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-09-reward-model-training https://hhy.homes/en/posts/hitchhiker-agentic-ai-09-reward-model-training Wed, 16 Sep 2026 00:10:00 GMT Reward models are the bridge between human preferences and the RL training signal.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-08-preference-optimization-variants https://hhy.homes/en/posts/hitchhiker-agentic-ai-08-preference-optimization-variants Wed, 16 Sep 2026 00:09:00 GMT This chapter covers the family of methods that extend or replace DPO with different objectives, data assumptions, or architectural trade-off… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-07-grpo-group-relative-policy-optimization https://hhy.homes/en/posts/hitchhiker-agentic-ai-07-grpo-group-relative-policy-optimization Wed, 16 Sep 2026 00:08:00 GMT Group Relative Policy Optimization (GRPO) [323] is a reinforcement learning algorithm designed specifically for language models that elimina… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-06-dpo-direct-preference-optimization https://hhy.homes/en/posts/hitchhiker-agentic-ai-06-dpo-direct-preference-optimization Wed, 16 Sep 2026 00:07:00 GMT PPO requires 4 models in memory (policy, reference, reward model, value head), complex RL infrastructure, and is notoriously unstable.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-05-ppo-proximal-policy-optimization https://hhy.homes/en/posts/hitchhiker-agentic-ai-05-ppo-proximal-policy-optimization Wed, 16 Sep 2026 00:06:00 GMT Problem: Vanilla policy gradient updates have no constraint on step size.… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-04-rl-foundations-for-language-models https://hhy.homes/en/posts/hitchhiker-agentic-ai-04-rl-foundations-for-language-models Wed, 16 Sep 2026 00:05:00 GMT Supervised fine-tuning (SFT) teaches a model to imitate demonstrations, but imitation has a ceiling: the model can never exceed the quality… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-03-introduction-to-reinforcement-learning https://hhy.homes/en/posts/hitchhiker-agentic-ai-03-introduction-to-reinforcement-learning Wed, 16 Sep 2026 00:04:00 GMT Reinforcement Learning (RL) is a paradigm where an agent learns to make sequential decisions by interacting with an environment, receiving r… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-02-systems-foundations-for-llms https://hhy.homes/en/posts/hitchhiker-agentic-ai-02-systems-foundations-for-llms Wed, 16 Sep 2026 00:03:00 GMT Modern large language models are trained and served almost exclusively on GPUs (Graphics Processing Units).… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-01-llm-architecture-and-optimization-methods https://hhy.homes/en/posts/hitchhiker-agentic-ai-01-llm-architecture-and-optimization-methods Wed, 16 Sep 2026 00:02:00 GMT This section covers the foundational architecture of large language models and the key optimization techniques that make training and infere… hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-0-preface https://hhy.homes/en/posts/hitchhiker-agentic-ai-0-preface Wed, 16 Sep 2026 00:01:00 GMT Preface to The Hitchhiker's Guide to Agentic AI: about the author, introduction and glossary of acronyms. hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/hitchhiker-agentic-ai-00-disclaimer https://hhy.homes/en/posts/hitchhiker-agentic-ai-00-disclaimer Wed, 16 Sep 2026 00:00:00 GMT Source notes, independence statement, AI usage disclosure and CC BY-SA 4.0 licensing for The Hitchhiker's Guide to Agentic AI. hitchhiker-agentic-ai craigmail.ai@gmail.com (Craig) https://hhy.homes/en/posts/say-hi-to-the-world https://hhy.homes/en/posts/say-hi-to-the-world Mon, 07 Sep 2026 00:00:00 GMT craigmail.ai@gmail.com (Craig)