English | MP4 | AVC 1920×1080 | AAC 44KHz 2ch | 57 Lessons (2h 20m) | 1.74 GB
The Complete Guide to Designing Modern AI Agents: Build Agentic Systems That Deliver Real-World ROI
AI Agents Are Redefining What Software Can Do. Are You Ready to Design Them?
If you’re fascinated by AI agents and want to truly understand how intelligent systems are architected, scaled, and controlled, then you are in exactly the right place.
AI Agents Design Blueprint: Architecting Agentic Systems is a comprehensive course that teaches you how modern AI agents actually work and how to design them for real-world use.
You’ll start with the foundations of agentic AI, clearly distinguishing between agentic workflows and full AI agents. From there, you’ll master the core building blocks of agentic systems, learn when to use agents (and when not to), and understand why architecture is the difference between a demo and a reliable system.
As the course progresses, you’ll go far beyond theory. You will learn:
- How to design robust prompting systems for agents
- How to implement agentic workflows such as routing, parallelization, orchestrator–worker, and evaluator–optimizer
- How to build single-agent and multi-agent architectures
- How to enable planning, decomposition, and reasoning
- How to design memory systems, including short-term context buffers, vector-based RAG, and entity-level memory
- How to ensure reliability, performance, and error recovery
- How to implement human-in-the-loop control, self-reflection, and oversight
- How to apply governance, guardrails, cost control, bias detection, and A/B shadow testing
- How to evaluate and benchmark AI agents properly in production
By the end of the course, you won’t just “use” AI agents—you’ll think like an AI agent architect, capable of designing systems that scale, adapt, and operate safely in real environments.
This course is ideal for:
- AI engineers and developers
- Data scientists and ML engineers
- Technical founders and product leaders
- Innovation managers and AI strategists
- Anyone who wants to move beyond chatbots into true agentic systems
Few technologies will reshape software, business operations, and entire industries as profoundly as AI agents. Companies are already shifting from simple LLM tools to agent-driven systems that act and learn.
If you want to stay ahead of this transformation, this course will give you the architecture playbook you need.
What Makes This AI Agents Course Different?
1. Production-Grade Architecture Focus
This is not just another “AI tools” course. We focus on design patterns, scalability, reliability, governance, and performance—the things that actually determine success in real products.
2. Instructor Excellence
Your instructor is Ned Krastev, CEO and founder of 365 Data Science, whose courses have educated over 5 million learners worldwide and are used by professionals and Fortune-level companies across the globe.
3. Visual, Story-Driven Learning
Complex agentic concepts are explained through high-quality animations, diagrams, and storytelling, making advanced architectures intuitive and easy to retain—far beyond traditional slide-based teaching.
Who this course is for:
- AI engineers and developers
- Data scientists and ML engineers
- Technical founders and product leaders
- Innovation managers and AI strategists
- Anyone who wants to move beyond chatbots into true agentic systems
Table of Contents
Introduction
1 What does the course cover
Foundations of Agentic AI
2 Agentic workflows vs AI agents
3 Core building blocks of an AI agent
4 When (and when not) to use AI agents
5 Why architecture matters Scaling reliability & control
Prompting for Agentic Systems
6 Section Introduction
7 General principles of prompt structuring
8 Prompting frameworks
9 Positive and negative prompting
10 Chain of Thought (CoT)
Agentic Workflows
11 Section Introduction
12 The augmented LLM
13 Prompt chaining
14 Routing
15 Parallelization
16 Orchestrator worker
17 Evaluator optimizer
Single Agent Architecture Patterns
18 Section Introduction
19 When to use a single agent architecture
20 Reflection
21 ReAct (Think Do)
22 Reflexion
Planning and Decomposition
23 Section Introduction
24 Task decomposition
25 The importance of planning
26 Plan and solve
27 ReWOO
28 Tree of thought
Multi Agent Architectures
29 Section Introduction
30 When to use single vs multi agent systems
31 Vertical vs horizontal architectures
32 Challenges with group conversations
33 Supervisor
34 Hierarchical teams
35 Dynamic teams
Execution Performance and Reliability
36 Section Introduction
37 Agents and asyncronous task execution
38 Performance metrics latency and cost
39 Error handling and recovery
Memory Systems
40 Section Introduction
41 Short term context buffer
42 Vector store RAG
43 Entity level memory
Oversight and Control
44 Section Introduction
45 Self reflection
46 Human in the loop
Governance and Safety
47 Governance patterns
48 Guardrails and policy еnforcement
49 Cost limiter
50 A B shadow testing
51 Bias and fairness systems
Evaluation and Benchmarking
52 Section Introduction
53 How to measure agent performance
54 Offline benchmarks
55 Online metrics
Conclusion
56 Conclusion
57 Bonus lecture
Resolve the captcha to access the links!
