Building and Optimizing AI Agent Workflows

Coursera via Coursera

Go to class Write review

Overview

Coursera Flash Sale

40% Off Coursera Plus for 3 Months!

Grab it

This long course equips you with practical knowledge and hands-on skills required to design, architect, and optimize autonomous AI agents that solve multi-step tasks reliably, efficiently, and responsibly. You will study reward-design and reinforcement-learning foundations to translate business objectives into robust reward signals, while learning to evaluate ethical, legal, and societal impacts of agent decision policies. The course covers competing reasoning-loop architectures (e.g., ReAct and Reflexion), modular agent component design with clear APIs, and search and planning strategies (A*, beam search, and heuristic augmentation). You will also practice feature engineering and model-interpretability methods to expose spurious correlations and produce explainable agent behaviors. Finally, the course guides you to make strategic modeling choices—such as fine-tuning large models versus training smaller task-specific models—and to package reproducible, reusable ML pipelines for agent subsystems. Throughout the course, practical labs and engineering-focused examples emphasize production-readiness, modularity, and trustworthiness.

Syllabus

Design Ethical AI Rewards and Policies

This module is for professionals and data scientists aiming to build responsible AI. As AI reshapes business, balancing performance with ethics is vital. This course provides a deep dive into reinforcement learning, teaching you to craft reward functions that align with corporate goals and global regulations like GDPR. Through hands-on labs and real-world case studies, you’ll learn to identify biases and implement fair governance. By bridging theory and practice, the program empowers you to lead initiatives that prioritize accountability, ensuring your AI systems deliver immense value without compromising integrity or public trust.

Architect Reusable AI Agent Systems

This module is for engineers transitioning from single-purpose bots to scalable, modular architectures. You’ll master advanced system design to build maintainable AI that evolves with business needs. The curriculum focuses on evaluating reasoning loops like ReAct and Reflexion through data-driven A/B testing. Through hands-on labs, you will apply software engineering best practices to develop reusable components—Planner, Memory, and Executor—using typed API contracts. By the end, you’ll be equipped to design and document a complete Python package of agent components, ready for seamless integration into high-value production environments.

Optimize Agentic AI: Algorithms for Peak Performance

This module is focused on building fast, scalable, and responsive systems. Recognizing that speed is as vital as intelligence, this program equips engineers to diagnose and resolve critical performance bottlenecks. You will master optimization techniques, replacing brute-force methods with sophisticated algorithms like beam search. Through hands-on labs, you’ll apply Big-O notation to analyze multi-tool reasoning pipelines and use profilers to pinpoint slowdowns. By learning to implement optimizations—such as indexing to reduce complexity from O(n^2) to O(log n)—you’ll gain the technical expertise to justify engineering decisions through professional proposals.

Hybrid AI Search Workflows

This module is for engineers and data scientists aiming to build intelligent, factually reliable search systems. While generative AI excels at reasoning, it often hallucinates; traditional search is accurate but lacks context. This program teaches you to architect hybrid workflows that ground LLMs with verifiable data. You will move beyond basic prompting to design and optimize systems for performance and cost. Through hands-on labs, you’ll master parameter tuning and modularizing code for production-ready CI/CD pipelines. By the end, you’ll be equipped to deploy trustworthy, context-aware AI applications that deliver reliable results at scale.

Engineer and Explain AI Model Decisions

This module is aimed for ML professionals who prioritize trust and accountability. In modern AI, high accuracy is insufficient; you must justify model outputs and mitigate harmful biases. This program teaches you to combine advanced feature engineering with model interpretability for ethical deployment. Through hands-on training, you will transform unstructured chat logs into model-ready tensors using Python, scikit-learn, TF-IDF, and embedding aggregation. You’ll then deconstruct "black box" models using SHAP to diagnose misclassifications and flag spurious correlations. By the end, you’ll develop an AI Model Decision Toolkit, equipping you to deliver stakeholder-ready reports that ensure transparent, reliable production AI.

Optimize AI: Build Reusable Model Pipelines

This is a module for engineers and data scientists focusing on scalable, maintainable workflows. Beyond simple model selection, this program teaches you to build standardized, reusable pipelines that accelerate development and ensure consistency. You will strategically evaluate trade-offs between large pre-trained models and efficient, custom alternatives, balancing performance with inference speed and cost. Through hands-on labs, you’ll master modular construction using Scikit-learn, emphasizing best practices for model management and versioning. By the end, you will transition from ad-hoc development to a systematic, pipeline-driven approach, essential for deploying robust, production-ready AI solutions.