Courses and Certifications

This two-hour course covers how to systematically assess and improve the performance of AI agents. It introduces methods for building agents from scratch, adding observability by collecting and visualizing traces, and using those traces to debug agent behavior. The course explains how to evaluate each component of an agent—such as routers, skills, and memory—by selecting appropriate evaluators (code-based, LLM-as-a-Judge, or human annotations), preparing test examples, and identifying relevant metrics. It also demonstrates how to structure evaluations into experiments to iterate on and enhance both the output quality and the reasoning paths taken by the agent. By the end, the course provides a framework for tracing, evaluating, and optimizing AI agents during development and in production.

This four-week intensive course is a hands-on guide for product managers ready to move from AI theory to shipped features. Designed for PMs familiar with concepts like RAG, evals, and prototyping tools, the program focuses on building and deploying agentic applications, mastering evaluations, and defining strategic roadmaps. Participants will prototype AI products, run evaluations with real tools like Cursor and LangGraph, and turn their learnings into actionable AI strategy docs. Led by Arize AI’s Head of Product — an alum of Spotify and Apple — the course includes live labs, peer feedback, and community access—equipping PMs to lead AI initiatives with confidence and credibility.

Created by the DataTalks.Club community, LLM Zoomcamp is a free, 10-week hands-on course for developers looking to build real-world AI applications using Retrieval-Augmented Generation (RAG). With step-by-step modules covering open-source LLMs, vector databases, orchestration, evaluation, and deployment, participants learn by doing—culminating in a portfolio-ready RAG project. The program includes homework, leaderboards, office hours, and an active Slack community to support your learning journey. Whether you’re new to AI or looking to apply LLMs in production, this course offers the structure, tools, and community to get started.

This course is designed to help developers, AI engineers, and technical product teams understand and implement autonomous, agentic AI workflows. Through hands-on modules and real-world scenarios, learners will explore key components like planning, memory, and tool integration, while mastering design patterns and architectural best practices for building robust, modular AI agents. The course also covers evaluation frameworks and deployment strategies, equipping participants with the skills to take agentic systems from prototype to production. This course provides the foundation to create scalable, intelligent agentic AI solutions.

In this three-part community course with Arize AI, you’ll dive deep into the fundamentals and frontiers of agentic AI systems. The series begins by unpacking the core components of AI agents—such as memory, planning, tool use, and system architectures — while contrasting single-agent and multi-agent designs. Through hands-on demos using Arize Phoenix, participants explore how to build and debug agents in practice. The second session tackles the unique challenges of evaluating agentic behavior, offering techniques like LLM-driven assessments, ground truth comparisons, and human feedback. Finally, the course advances into cutting-edge methods such as path-based reasoning, convergence tracking, and peer evaluation, equipping attendees with the tools to assess not just what an agent does—but how and why it made those decisions.

Dive into the world of AI agent development in this comprehensive bootcamp. You’ll gain hands-on experience with the latest tools and techniques in AI agent development. From understanding core architectures to troubleshooting complex issues, our expert-led sessions will equip you with the knowledge and skills to build sophisticated AI agents that can tackle real-world challenges. You’ll learn how to navigate popular frameworks, optimize performance, and evaluate your agents effectively.