Advanced Agentic AI: Self-Improving Systems & Frameworks

Packt via Coursera

Go to class Write review

Overview

Google, IBM & Meta Certificates – 40% Off

One Coursera Plus subscription covers most Professional Certificates on Coursera.

Unlock All Certificates

Advance your expertise in agentic AI by exploring self-improving systems, strategic implementation roadmaps, and practical frameworks for deploying production-grade multi-agent solutions. Learn from detailed use cases and comparative analyses of leading agent frameworks. This course guides learners through advanced adaptation techniques, operational frameworks for self-improving agents, and strategic roadmaps for implementing agentic patterns at scale. Participants will examine real-world use cases in loan processing, compare leading agent frameworks such as CrewAI and LangGraph, and gain insights into responsible AI practices and production readiness. The course concludes with actionable strategies for achieving higher levels of agentic maturity and operational excellence. Combining in-depth case studies, comparative framework analyses, and strategic guidance, the course empowers learners to design, implement, and iterate on advanced agentic AI systems. Emphasis is placed on practical application, maturity assessment, and continuous improvement. This course is part three of a three-course Specialization designed to build a complete and cohesive understanding of the subject. While it offers valuable skills on its own, you'll gain the most benefit by progressing through all three courses as a structured learning journey. This course is based on Agentic Architectural Patterns for Building Multi-Agent Systems, by Dr. Ali Arsanjani and Juan Pablo Bustos. Packt is one of the world's most prolific publishers of cutting-edge technical content. For over two decades we've made it our mission to curate and publish the knowledge of only the very best technical experts. We focus on real-world courses that help our customers get the job done, with coverage that extends across a wide range of established and cutting-edge technical topics. If you're an individual or an organisation that embraces learning by doing, Packt is the perfect fit for you.

Syllabus

Advanced Adaptation: Building Agents That Learn
- This module delves into advanced strategies for building self-improving agentic systems, focusing on operational frameworks, feedback-driven learning, and robust evaluation methods. Learners will explore the R⁵ model, preference-controlled data generation, advanced tuning patterns, and techniques for measuring both technical and business value. Practical guidance on cost management and adversarial testing ensures that agents are not only adaptive but also production-ready and efficient.
A Practical Roadmap: Implementing Agentic Patterns by Maturity Level
- This module guides learners through a step-by-step roadmap for implementing agentic AI systems, progressing from foundational prototypes to advanced, self-improving ecosystems. Learners will explore key architectural decisions, critical design patterns, and strategies for aligning project goals with system maturity. Practical tools and reflection guides are provided to help translate theory into actionable project plans.
Use Case: A Single Agent for Loan Processing
- This module guides learners through the implementation of a monolithic loan processing agent using the Fractal Chain of Thought (FCoT) framework. Participants will design, configure, and test the agent in a Colab environment, analyze its decision-making process, and evaluate its performance and limitations in real-world scenarios.
Use Case: A Multi-Agent System for Loan Processing
- This module guides learners through the transition from a single-agent to a multi-agent system for loan processing, emphasizing hierarchical architectures and the Supervisor pattern. Learners will explore the design and implementation of specialized agents, contextual state management, and advanced patterns like FCoT to enhance scalability and resilience. The module also discusses future directions in agentic collaboration.
Agent Frameworks – Use Case: A Multi-Agent System for Loan Processing with CrewAI and LangGraph
- This module guides learners through the practical use of agent frameworks such as ADK, CrewAI, and LangGraph for building multi-agent AI systems. You will compare their architectures, explore real-world implementation strategies for loan processing, and consider best practices for observability and responsible AI. By the end, you'll be equipped to select and apply the most suitable framework for your own agentic applications.
Conclusion: Charting Your Agentic AI Journey
- This module guides learners through the practical steps of implementing, scaling, and managing agentic AI systems using large language models. You will review key concepts, explore strategies for organizational maturity, and learn how to transition from experimentation to robust, production-ready deployments.