← на главную

> Карта сайта

узлов: 160 · связей: 328 — клик по узлу открывает заметку

освоение считается локально (прочитанное светится)

00-Index / Agent = LLM + Context + Tools Agent = LLM + Context 00-Theory / Glossary Glossary 01-Agent-Fundamentals / Harness Engineering Harness Engineering 01-Agent-Fundamentals / Agent Loop in Code Agent Loop in Code 01-Agent-Fundamentals / Tool Calling Tool Calling 01-Agent-Fundamentals / Context as the Agent’s Eyes Context as the Agent’s 01-Agent-Fundamentals / Guardrails Guardrails 01-Agent-Fundamentals / Orchestration - Workflow vs. Autonomous Agent Orchestration - Workfl 01-Agent-Fundamentals / ACI Design ACI Design 01-Agent-Fundamentals / Observation and Action Spaces Observation and Action 01-Agent-Fundamentals / Five Categories of Tools Five Categories of Too 01-Agent-Fundamentals / Five Functions of Harness Five Functions of Harn 01-Agent-Fundamentals / Agent Architecture — Summary Agent Architecture — S 01-Agent-Fundamentals / Three Levels of Capability Update Three Levels of Capabi 01-Agent-Fundamentals / Agent Formula Agent Formula 01-Agent-Fundamentals / ReAct Cycle ReAct Cycle 01-Agent-Fundamentals / Evolution of Engineering Paradigms Evolution of Engineeri 01-Telegram / GemStuffer GemStuffer 01-Telegram / Does it work? Does it work? 01-Telegram / How Telegram Bridge Works How Telegram Bridge Wo 01-Telegram / 🤖 AI Interns Went for a Walk: How OpenAI Bots Hacked Hugging Face 🤖 AI Interns Went for 02-Context-Engineering / Agent Skills Agent Skills 02-Context-Engineering / Chat Template Chat Template 02-Context-Engineering / KV Cache KV Cache 02-Context-Engineering / Prompt Cache Prompt Cache 02-Context-Engineering / Context Engineering Context Engineering 02-Context-Engineering / Prompt Engineering Prompt Engineering 02-Context-Engineering / Prompt Injection Prompt Injection 02-Context-Engineering / Five Context Components Five Context Component 02-Context-Engineering / Context Compression Context Compression 02-Context-Engineering / Agent Status String Agent Status String 02-Context-Engineering / Three Rules of KV Cache Friendliness Three Rules of KV Cach 02-Context-Engineering / Four Message Roles API Four Message Roles API 02-Обучение / Learning Learning 03-Memory-Knowledge-Base / GraphRAG GraphRAG 03-Memory-Knowledge-Base / RAG RAG 03-Memory-Knowledge-Base / RAPTOR RAPTOR 03-Memory-Knowledge-Base / User as Code User as Code 03-Memory-Knowledge-Base / Agentic RAG Agentic RAG 03-Memory-Knowledge-Base / Hybrid Search Hybrid Search 03-Memory-Knowledge-Base / Memory Hierarchy Memory Hierarchy 03-Memory-Knowledge-Base / Cognitive Types of Memory Cognitive Types of Mem 03-Memory-Knowledge-Base / Contextual Search Contextual Search 03-Memory-Knowledge-Base / Dense Embedding Dense Embedding 03-Memory-Knowledge-Base / Document Splitting Document Splitting 03-Memory-Knowledge-Base / BM25 Sparse Retrieval BM25 Sparse Retrieval 03-Memory-Knowledge-Base / Structured Indexing Structured Indexing 03-Memory-Knowledge-Base / Three Levels of Memory Evaluation Three Levels of Memory 03-Memory-Knowledge-Base / User Memory Formats User Memory Formats 04-Tools / Balancing Tool Granularity Balancing Tool Granula 04-Tools / Dynamic Tool Loading Dynamic Tool Loading 04-Tools / Parameter Passing Fidelity Parameter Passing Fide 04-Tools / The Art of Tool Description The Art of Tool Descri 04-Tools / Tool Classification Tool Classification 04-Tools / Sidecar Mechanism Sidecar Mechanism 04-Tools / Proposer-Reviewer Proposer-Reviewer 04-Tools / Interrupts and Placeholders Interrupts and Placeho 04-Tools / Proactive Tool Discovery Proactive Tool Discove 04-Tools / MCP Protocol MCP Protocol 04-Tools / MCP Security Risks MCP Security Risks 04-Tools / Event-Driven Asynchronous Agent Event-Driven Asynchron 04-Tools / Specialized Tool vs Skill Specialized Tool vs Sk 04-Tools / Human in the Loop Human in the Loop 04-Блог / Blog Blog 05-Coding-Agent / Sessionless Sessionless 05-Coding-Agent / Generative UI Generative UI 05-Coding-Agent / Code as a Tool for Thought Code as a Tool for Tho 05-Coding-Agent / Code as Meta-Ability Code as Meta-Ability 05-Coding-Agent / Code as Business Rules Constraint Code as Business Rules 05-Coding-Agent / Agent Loyalty Agent Loyalty 05-Coding-Agent / Filesystem Paradigm for Knowledge Filesystem Paradigm fo 05-Coding-Agent / Agent Self-Bootstrapping Agent Self-Bootstrappi 05-Coding-Agent / Failures and Recovery Failures and Recovery 05-Coding-Agent / Semantic Dissection of Commands Semantic Dissection of 05-Coding-Agent / Seven Basic Tools Seven Basic Tools 05-Coding-Agent / Deadly Triad Deadly Triad 05-Coding-Agent / File System as Agent Core File System as Agent C 05-О-проекте / About the Project About the Project 06-Evaluation / LLM-as-a-Judge LLM-as-a-Judge 06-Evaluation / Agent Cost Analysis Agent Cost Analysis 06-Evaluation / Agent Evaluation Metrics Agent Evaluation Metri 06-Evaluation / Agent Observability Agent Observability 06-Evaluation / From Benchmark to Improvements From Benchmark to Impr 06-Evaluation / Simulation Environment Simulation Environment 06-Evaluation / Evaluation Environments Evaluation Environment 06-Evaluation / Statistical Significance of Evaluation Statistical Significan 06-Evaluation / Three Levels of the Evaluation System Three Levels of the Ev 06-Evaluation / Four Principles of a Rubric Four Principles of a R 07-Post-Training / GRPO GRPO 07-Post-Training / KL - mode-seeking vs mass-covering KL - mode-seeking vs m 07-Post-Training / On-Policy Distillation On-Policy Distillation 07-Post-Training / RLHF RLHF 07-Post-Training / RLVR and Verifiable Rewards RLVR and Verifiable Re 07-Post-Training / SFT Memorizes - RL Generalizes SFT Memorizes - RL Gen 07-Post-Training / Data and Environment Matter More Than the Algorithm Data and Environment M 07-Post-Training / Reward for Process and Reward for Outcome Reward for Process and 07-Post-Training / OPSD Self-Distillation OPSD Self-Distillation 07-Post-Training / SFT First, Then RL SFT First, Then RL 07-Post-Training / Three Stages of Post-Training Three Stages of Post-T 08-Continuous-Evolution / Safety Boundaries of Self-Modification Safety Boundaries of S 08-Continuous-Evolution / Two Loops of Evolution Two Loops of Evolution 08-Continuous-Evolution / Continuous Agent Evolution Continuous Agent Evolu 08-Continuous-Evolution / Prompt and Skill Update Prompt and Skill Updat 08-Continuous-Evolution / Model Parameter Updates Model Parameter Update 08-Continuous-Evolution / Learning Signals from Trajectories Learning Signals from 08-Continuous-Evolution / Dream Learning Dream Learning 08-Continuous-Evolution / From Experience to Knowledge From Experience to Kno 08-Continuous-Evolution / From Experience to Programs From Experience to Pro 08-Continuous-Evolution / Check - Release - Rollback Check - Release - Roll 08-Continuous-Evolution / Four Update Carriers Four Update Carriers 09-Multimodality / Computer Use Computer Use 09-Multimodality / Sim2Real Sim2Real 09-Multimodality / VLA Models VLA Models 09-Multimodality / Architecture of Thinking: Fast and Slow Architecture of Thinki 09-Multimodality / Visual GUI Localization Visual GUI Localizatio 09-Multimodality / Voice Paradigms Voice Paradigms 09-Multimodality / Two-Level Architecture: Planning-Control Two-Level Architecture 09-Multimodality / Cascaded Voice Pipeline Cascaded Voice Pipelin 09-Multimodality / Full Duplex Full Duplex 09-Multimodality / Streaming Speech Perception Streaming Speech Perce 09-Multimodality / Human-like Speech Synthesis Human-like Speech Synt 10-Multi-Agent / Decentralized Model and Handoff Decentralized Model an 10-Multi-Agent / Requesting Status and Trajectory Persistence Requesting Status and 10-Multi-Agent / Inter-Agent Communication Inter-Agent Communicat 10-Multi-Agent / Orchestration Model Orchestration Model 10-Multi-Agent / Multi-Agent Role Shifting Multi-Agent Role Shift 10-Multi-Agent / Multi-Agent Interaction Without Shared Context Multi-Agent Interactio 10-Multi-Agent / Multi-Agent vs. Single Agent Multi-Agent vs. Single 10-Multi-Agent / Society of Agents Society of Agents 10-Multi-Agent / Shared vs. Separate Context Shared vs. Separate Co 10-Multi-Agent / A2A Protocol A2A Protocol 10-Multi-Agent / Failure Modes of Multi-Agent Systems Failure Modes of Multi 10-Multi-Agent / Interaction Topologies Interaction Topologies 10-Multi-Agent / File System in Multi-Agent Systems File System in Multi-A 10-Multi-Agent / Emergent Behavior of Agents Emergent Behavior of A 11-Afterword / Harness is Gradually Absorbed by the Model Harness is Gradually A 11-Afterword / Two Clouds of Agent Heaven Two Clouds of Agent He 11-Afterword / Model-Agent Flywheel Model-Agent Flywheel 11-Afterword / Model and Agent Co-evolution Model and Agent Co-evo Дайджест / Digest Digest Дайджест / AI Interns Gone Wild: A Simple Explanation of One Hack AI Interns Gone Wild: Курс-агент-с-нуля / 00. About the Course 00. About the Course Курс-агент-с-нуля / Lesson 1. Agent Formula Lesson 1. Agent Formul Курс-агент-с-нуля / Lesson 2. The ReAct Cycle Lesson 2. The ReAct Cy Курс-агент-с-нуля / Lesson 3. Harness and Guardrails Lesson 3. Harness and Курс-агент-с-нуля / Lesson 4. Tools Lesson 4. Tools Разборы / Analyses Analyses Разборы / When Agents Went Hunting: A Breakdown of the OpenAI / Hugging Face Incident When Agents Went Hunti Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / I. Introduction — Summary I. Introduction — Summ Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / II. OpenAI Evaluation Environment – Summary II. OpenAI Evaluation Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / III. Pre-Incident Actions – Summary III. Pre-Incident Acti Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / IV. Hugging Face Intrusion – Executive Summary IV. Hugging Face Intru Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / V. Attack on OpenAI Infrastructure – Summary V. Attack on OpenAI In Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / VI. Initial Response — Summary VI. Initial Response — Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / VII. Safety Lessons – Takeaways VII. Safety Lessons – Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / VIII. Alignment Lessons – Takeaways VIII. Alignment Lesson Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / IX. OpenAI Action Plan – Summary IX. OpenAI Action Plan Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / X. Key Technical Events – Summary X. Key Technical Event Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / Table of Contents — OpenAI Technical Report on the Hugging Face Incident Table of Contents — Op Технический_отчёт_OpenAI_по_инциденту_Hugging_Face / OpenAI Technical Report on the Hugging Face Incident OpenAI Technical Repor
▸ 00-Index▸ 00-Theory▸ 01-Agent-Fundamentals▸ 01-Telegram▸ 02-Context-Engineering▸ 02-Обучение▸ 03-Memory-Knowledge-Base▸ 04-Tools▸ 04-Блог▸ 05-Coding-Agent▸ 05-О-проекте▸ 06-Evaluation▸ 07-Post-Training▸ 08-Continuous-Evolution▸ 09-Multimodality▸ 10-Multi-Agent▸ 11-Afterword▸ Дайджест▸ Курс-агент-с-нуля▸ Разборы▸ Технический_отчёт_OpenAI_по_инциденту_Hugging_Face
// достижения

считается локально: заметка прочитана = 12+ секунд на странице; раздел освоен = прочитаны все его заметки