Further Reading | The Agentic Wiki

A curated collection of resources for deepening your understanding of agentic AI systems, their design patterns, and their practical applications.

Primary Source

Building Effective Agents --- Anthropic (2024). The foundational reference for this wiki. Presents a practical taxonomy of agentic patterns, from augmented LLMs through workflows to fully autonomous agents, with clear guidance on when to use each.

ReAct: Synergizing Reasoning and Acting in Language Models --- Yao et al. (2022). Introduces the ReAct paradigm where LLMs interleave reasoning traces and task-specific actions, forming the conceptual basis for many modern agent architectures.
Toolformer: Language Models Can Teach Themselves to Use Tools --- Schick et al. (2023). Demonstrates how language models can learn to use external tools autonomously, a key capability underlying agentic tool use.
Reflexion: Language Agents with Verbal Reinforcement Learning --- Shinn et al. (2023). Presents a framework for agents that improve through self-reflection, directly relevant to evaluator-optimizer patterns.
Voyager: An Open-Ended Embodied Agent with Large Language Models --- Wang et al. (2023). Explores an LLM-powered agent that continuously explores, learns skills, and makes discoveries in open-ended environments, demonstrating long-horizon autonomous behavior.
Chain-of-Thought Prompting Elicits Reasoning in Large Language Models --- Wei et al. (2022). The foundational paper on chain-of-thought prompting, a technique that underpins reasoning in virtually all agentic systems.

The Shift from Models to Compound AI Systems --- Berkeley AI Research (2024). Argues that the future of AI lies in systems that combine multiple models, retrievers, and tools rather than monolithic models alone.
What We Learned from a Year of Building with LLMs --- O’Reilly (2024). Practical lessons from production LLM applications, covering prompting, RAG, agent design, and operational concerns.
LLM Powered Autonomous Agents --- Lilian Weng (2023). A comprehensive survey of LLM-based agent architectures covering planning, memory, and tool use.
Patterns for Building LLM-based Systems & Products --- Eugene Yan (2023). Practical patterns for LLM systems including evaluation, RAG, fine-tuning, and agent design, with emphasis on production readiness.

Generative AI on AWS --- Chris Fregly, Antje Barth, Shelbee Eigenbrode (O’Reilly, 2024). Covers building generative AI applications including agent architectures, with practical implementation guidance.
Building LLMs for Production --- Louis-Fran\u00e7ois Bouchard, Louie Peters (2024). Addresses the engineering challenges of taking LLM-based systems, including agents, from prototype to production.