Nokku AI course · Certificateintermediate
Building AI Agents: A Hands-On Guide for Developers
A hands-on Python course for developers who want to build reliable AI agents. Starting from a single model call, you'll build a tool-calling agent loop, then add well-designed tools, retrieval with citations, memory, an evaluation harness, prompt-injection guardrails and cost controls, and ship a production-ready customer-support agent. Every code sample is a real, tested file.
- Chapters
- 10 chapters
- Duration
- 4h 9m
- Level
- intermediate
What you'll learn
- Call language model APIs correctly, including conversations, streaming and stop reasons
- Implement tool calling and a robust agent loop with step limits, error handling and hooks
- Design agent tools with clear descriptions, concise outputs, validation and approval gates
- Ground agents in your own data with retrieval and citations
- Manage context and long-term memory safely
- Build an evaluation harness with code checks and a calibrated model judge
- Defend against prompt injection with layered guardrails enforced in code
- Measure and control cost and latency with budgets, prompt caching and model routing
- Assemble, evaluate and launch a production support agent in stages
Course content
10 chapters · 4h 9m total · final assessment
- 1Calling a Model: The Foundations23 min
- 2Tool Calling From Scratch17 min
- 3The Agent Loop24 min
- 4Designing Tools Agents Can Use26 min
- 5Retrieval: Grounding Agents in Your Data23 min
- 6Memory and Context Management23 min
- 7Evaluating Agents in Code25 min
- 8Guardrails and Prompt-Injection Defence24 min
- 9Cost, Latency and Model Routing26 min
- 10Capstone: Shipping a Production Support Agent38 min
- Final assessment & certificate
How it works
- 01
Read each chapter
Short, focused lessons you can finish in one sitting.
- 02
Pass the quizzes
Score 70% or more on a chapter's quiz to unlock the next one.
- 03
Earn your certificate
Pass the final assessment to get a certificate with a verification link.