Course Information

  • Sessions 2 days
  • Duration 15 hrs
  • Level Beginner
  • Assessment NA

Venue

Course Brochure

Prerequisite

This course is for beginners. No prerequisite is required.

Certification

  • Certificate of Completion from Tertiary Courses - Upon meeting at least 75% attendance and passing the assessment(s), participants will receive a Certificate of Completion from Tertiary Courses.

Develop Multi-Agent Systems with OpenAI Agent SDK

Course Code: C765
Share

What's This Course About

Develop Multi-Agent Systems with OpenAI Agent SDK teaches you how to design, build and deploy multi-agent AI systems using OpenAI's Agents SDK. Multi-agent systems split complex work across specialised agents that use tools, retrieve knowledge, hand off tasks to one another and work under guardrails, enabling applications that are more capable and reliable than a single chatbot. The course is designed for developers, AI engineers, data scientists and technical professionals with Python experience who want to build production-ready agentic applications.

You will start with an introduction to the OpenAI Agents SDK, covering agents, instructions, tools, handoffs, guardrails, sessions and tracing, and building your first agents in Python. Next you will build a retrieval-augmented generation (RAG) agent that answers questions accurately from your own documents using file search and vector stores. You will then orchestrate multiple agents, designing triage and specialist agents, managing handoffs and parallel work, and tracing runs to debug and improve agent behaviour.

The course continues with the Model Context Protocol (MCP), connecting your agents to external tools and data sources through MCP servers, and closes with fine-tuning LLM models, learning when and how to customise a model for your domain and evaluate the results. Throughout the course you will also cover safety, guardrails, cost control and deployment practices. Every topic is hands-on. By the end, you will be able to develop, evaluate and deploy multi-agent systems with the OpenAI Agents SDK confidently.

This course is part of our AI training in Ghana, where you can browse all our AI courses. See more multi AI agents courses.

Why Take This Course?

  • Hands-On Learning Learn by doing. Every topic is practised hands-on, so you leave with skills you can use at work straight away.
  • Live Coaching Experienced instructors coach you live, answer your questions and guide you step by step.
  • Free Post-Course Consultation Stuck after class? Message your trainer any time for guidance as you apply what you have learned.
  • Always-Updated Materials Materials are refreshed to stay relevant. Log in to our LMS any time to get the latest version.

Course Fee

₵3,000.00

Course Date

Course Time

Additional Note

Please bring your own laptop for hands-on training. If you don't have laptop, we can provide spare laptop for training use.

Post-Course Support

  • We provide free consultation related to the subject matter after the course.
  • Please email your queries to siraj@tertiarycourses.com.gh and we will forward your queries to the subject matter experts and get back to you as soon as possible.

Cancellation & Reschedule Policy

  • We reserve the right to cancel or re-schedule the course due to unforeseen circumstances. If the course is cancelled, we will refund 100% to participants.
  • Note: the venue of the training is subject to changes due to class size and availability of the classroom. The minimum class size to start a class is 3 Pax.

Course Details

Course Details

What You'll Learn

Topic 1: Introduction to OpenAI Agent Development Kit

Topic 2: Building RAG Agent

Topic 3: Orchestrating Multiple Agents

Topic 4: Model context protocol (MCP)

Topic 5: Fine Tuning LLM Models

Course Info

Minimum Entry Requirement

Knowledge and Skills

  • Able to operate using computer functions
  • Minimum 3 GCE ‘O’ Levels Passes including English or WPL Level 5 (Average of Reading, Listening, Speaking & Writing Scores)

Attitude

  • Positive Learning Attitude
  • Enthusiastic Learner

Experience

  • Minimum of 1 year of working experience.

Target Age Group: 18-65 years old

Minimum Software/Hardware Requirement

You will need the following for this course:

Install and test the listed software before class. Confirm the required version and account or licence access in your joining instructions. Cloud and AI services may charge subscription or usage fees.

Also prepare the tools needed for the practical exercises:

Confirm the version, account permissions and any API usage budget with the trainer before class.

Bring a Windows or Mac laptop supported by the listed software, with a working keyboard and sufficient storage for course exercises.

Bring your own computer and charger, with a reliable internet connection and permission to install the required software. For live online classes, also prepare a microphone or headset.

If you need help arranging a suitable device, contact us before the class to confirm availability.

Job Roles

Job Roles

  • NLP (Natural Language Processing) Engineer
  • Machine Learning Engineer
  • AI Application Developer
  • Chatbot Developer
  • AI Research Scientist
  • AI Product Manager
  • Data Scientist specializing in text data
  • Conversational AI Designer
  • AI Integration Specialist
  • Language Model Trainer
  • Content Personalization Engineer
  • AI Platform Engineer
  • Cognitive Computing Specialist
  • AI Technical Consultant
  • AI Software Architect

Trainers

Trainers

Dr. Muhammed Siraj is a cybersecurity specialist and academic with 15 years of experience, serving as Senior Lecturer in the Department of Informatics and Director of Graduate and Continuous Education at Academic City University in Accra. He holds a PhD in Security in Computing from Universiti Putra Malaysia, where his research focused on blockchain security, an MSc in Information and Communication Technology from Aalborg University in Denmark, and a degree from the University of Ghana. He is a licensed independent assessor with Ghana's Cyber Security Authority, holds leadership roles in national cybersecurity and privacy professional bodies, and is certified as a Certified Ethical Hacker (CEH), Computer Hacking Forensic Investigator (CHFI) and Certified Data Protection Supervisor (CDPS). Before joining Academic City, he served as Acting Director of ICT and Data Protection Supervisor at Wisconsin International University College. His expertise spans penetration testing, digital forensics, incident response, malware analysis, cryptography, cloud and IoT security, zero trust architecture, cyber threat intelligence, data protection, and governance, risk and compliance.

Dr. Siraj's research interests cover blockchain, cybersecurity, distributed computing, consensus protocols and machine learning, with published work on blockchain consensus security, smart-contract privacy preservation for sensitive records in the cloud, and new consensus mechanisms. This background gives him a distinctive perspective on artificial intelligence: as organisations adopt AI tools, agents and data-driven automation, he helps learners understand how to protect the data these systems depend on, recognise new attack surfaces, and put sound governance, privacy and access controls in place so that AI can be used safely and responsibly. In the classroom he connects technical practice with regulation, privacy and human factors, drawing on real-world security and data-protection cases to show how concepts apply in professional settings, and his experiential, hands-on teaching style has made him a trusted trainer for students and working professionals alike.

Multi-agent systems, retrieval-augmented generation and tool integration are the building blocks Dr. Alfred Ang works with every week. Dr. Ang is a full-stack AI developer, Agentic AI solution architect and corporate trainer, and serves as CTO of Tertiary Infotech Academy Pte Ltd, the Singapore training organisation behind Tertiary Courses. Dr. Ang's RAG work includes n8n-ragdemo, in which uploaded PDFs feed a vector store that a Telegram AI agent queries, and n8n-ragdemo2, a customer-service RAG chatbot over a library of course brochures. Dr. Ang has also built n8n-hrportal, which wraps an OpenAI HR chatbot in input and output guardrails with human-in-the-loop approval, and digitalhuman, a voice-cloned avatar trained on custom content through RAG. On the orchestration side, an ERP prototype coordinates several AI agents for purchase orders, invoice matching and replenishment, hermes-video-pipeline lets an agent produce YouTube videos end-to-end, and AI-LMS-TMS embeds agent automation into a full training management system. Much of this was developed with Claude Code and the Claude Agent SDK, using Model Context Protocol servers to connect agents to tools and data. More than 300 public repositories at github.com/alfredang show the practical side of agent engineering: designing roles, passing context safely, handling failures and deploying agents that keep working once real users arrive. Dr. Ang also pays close attention to cost, latency and observability, because a multi-agent system that is slow, expensive or impossible to debug rarely survives contact with real users.

Develop Multi-Agent Systems with OpenAI Agent SDK turns that experience into hands-on practice. The course opens with an introduction to the OpenAI agent development kit, covering agents, tools, handoffs and guardrails, and setting up a working Python project. Learners then build a RAG agent, chunking and embedding documents, storing them in a vector database and grounding answers in retrieved sources. The third topic orchestrates multiple agents, assigning specialist roles, routing requests and passing context between agents to complete multi-step tasks. The fourth topic introduces the Model Context Protocol, showing how MCP servers expose tools and data to agents in a standard, reusable way. The final topic covers fine-tuning LLM models, including when fine-tuning beats prompting or retrieval, how to prepare training data and how to evaluate the result. Dr. Ang codes live and explains design choices as they arise, drawing on production agents to show common pitfalls such as runaway tool calls, context overflow and hallucinated citations, and how to guard against them. Participants build and extend their own multi-agent application throughout the course, leaving with working code they can adapt for business use. Developers, data scientists and technical architects with basic Python skills will gain a practical, end-to-end view of agent engineering. Learners also explore tracing, logging and evaluation techniques for agents, so they can see exactly which agent made each decision and measure quality as the system grows. To explore the RAG chatbots and agent systems behind these lessons, visit Dr. Ang's portfolio at https://github.com/alfredang. Connect with Dr. Alfred Ang on LinkedIn at https://www.linkedin.com/in/angchewhoe/

Review

Customer Reviews (6)

Excellent course Review by Course Participant/Trainee
1. Do you find the course meet your expectation?
2. Do you find the trainer knowledgeable in this subject?
3. How do you find the training environment
The trainer was knowledgeable and explained the concepts clearly with real-world examples. (Posted on 7/22/2024)

Items 6 to 6 of 6 total

per page
Page:
  1. 1
  2. 2

Write Your Own Review

You're reviewing: Develop Multi-Agent Systems with OpenAI Agent SDK

How do you rate this product? *

  1 star 2 stars 3 stars 4 stars 5 stars
1. Do you find the course meet your expectation?
2. Do you find the trainer knowledgeable in this subject?
3. How do you find the training environment