Course Outline
Day 1
Anatomy of a Modern AI Agent
Understanding agents as autonomous reasoning and acting systems, beyond simple chatbots.
Exploring reactive, proactive, hybrid, and goal-directed agent paradigms.
Identifying core components: perception, planning, memory, tool utilization, and action.
Evaluating design tradeoffs between single-agent and multi-agent architectures.
Agent Frameworks and the Modern Stack
Analyzing LangChain, LlamaIndex, AutoGen, and CrewAI, including their respective tradeoffs.
Comparing modern frameworks with classical approaches like JADE and SPADE.
Selecting the appropriate framework based on specific production requirements.
Mastering tool calling, function calling, and generating structured outputs.
Hands-on: Scaffolding a single Python agent equipped with tool calls.
Multi-Agent System Architectures
Examining centralized, decentralized, hybrid, and layered multi-agent system (MAS) designs.
Understanding FIPA ACL, message-passing mechanisms, and their modern equivalents.
Studying coordination patterns such as planning, negotiation, and synchronization.
Investigating emergent behavior and self-organization within agent populations.
Decision-Making and Learning in Agents
Applying game theory to cooperative and competitive agent interactions.
Implementing reinforcement learning within multi-agent environments.
Facilitating transfer learning and knowledge sharing across agents.
Resolving conflicts and establishing trust among coordinating agents.
Day 2
Multi-Modal Foundations for Agents
Viewing multi-modal AI as a unified workflow encompassing text, image, speech, and video.
Reviewing leading multi-modal models: GPT-4 Vision, Gemini, Claude, and Whisper.
Exploring fusion techniques for integrating modalities within an agent's reasoning loop.
Balancing latency, cost, and accuracy in multi-modal pipelines.
Building the Perception Layer
Utilizing image processing for agents, including classification, captioning, and object detection.
Implementing speech recognition via Whisper ASR and streaming transcription.
Generating natural voice interactions through text-to-speech synthesis.
Linking perception outputs to LLM-driven reasoning and tool selection.
Hands-On - Building a Multi-Modal Agent in Python
Defining the agent’s task, context window, and tool inventory.
End-to-end integration of GPT-4 Vision and Whisper APIs.
Implementing memory, state management, and conversation handling.
Safely adding tool calls that generate real-world side effects.
Hands-On - Orchestrating a Multi-Agent System
Composing specialized agents using AutoGen or CrewAI.
Defining roles, responsibilities, and inter-agent communication protocols.
Managing resource allocation and coordination in a simulated environment.
Logging agent reasoning, tool calls, and decisions for inspection and audit.
Day 3
Threat Surface of Production AI Agents
Identifying what makes agentic AI uniquely vulnerable compared to traditional software.
Mapping the attack surface: data, model, prompt, tool, output, and interface layers.
Conducting threat modeling for agent-based systems with autonomous tool use.
Comparing AI cybersecurity practices with traditional cybersecurity standards.
Adversarial Attacks Hands-On
Examining adversarial examples and perturbation methods: FGSM, PGD, DeepFool.
Distinguishing between white-box and black-box attack scenarios.
Analyzing model inversion and membership inference attacks.
Understanding data poisoning and backdoor injection during training.
Addressing prompt injection, jailbreaking, and tool misuse in LLM-based agents.
Defensive Techniques and Model Hardening
Employing adversarial training and data augmentation strategies.
Utilizing defensive distillation and other robustness techniques.
Applying input preprocessing, gradient masking, and regularization.
Implementing differential privacy, noise injection, and managing privacy budgets.
Leveraging federated learning and secure aggregation for distributed training.
Hands-On with the Adversarial Robustness Toolbox
Simulating attacks against the multi-modal agent developed on Day 2.
Measuring robustness under perturbation and quantifying performance degradation.
Iteratively applying defenses and re-evaluating attack success rates.
Stress-testing tool-call pathways and prompt injection vectors.
Day 4
Risk Management Frameworks for AI
Implementing the NIST AI Risk Management Framework: govern, map, measure, manage.
Navigating ISO/IEC 42001 and emerging AI-specific standards.
Mapping AI risks to existing enterprise GRC frameworks.
Meeting AI accountability, auditability, and documentation requirements.
Regulatory Compliance for Agentic Systems
Understanding the EU AI Act: risk tiers, prohibited uses, and obligations for high-risk systems.
Assessing GDPR and CCPA implications for agent data pipelines.
Reviewing the U.S. Executive Order on Safe, Secure, and Trustworthy AI.
Applying sector-specific guidance for finance, healthcare, and public services.
Managing third-party risk and supplier AI tool usage.
Ethics, Bias, and Explainability
Detecting and mitigating bias across agent perception and reasoning.
Recognizing explainability and transparency as critical security properties.
Ensuring fairness, minimizing downstream harm, and promoting responsible deployment.
Designing inclusive and auditable agent behaviors.
Production Deployment, Monitoring, and Incident Response
Adopting secure deployment patterns for single and multi-agent systems.
Implementing continuous monitoring for drift, anomalies, and abuse.
Establishing logging, audit trails, and forensic readiness for agent actions.
Developing AI security incident response playbooks and recovery procedures.
Analyzing case studies of real-world AI breaches and extracting lessons learned.
Capstone and Synthesis
Reviewing the multi-modal multi-agent system constructed throughout the course.
Conducting an end-to-end pipeline review: design, build, secure, govern, deploy.
Self-assessing the system against NIST AI RMF functions.
Looking ahead to emerging trends in agentic AI and AI security.
Summary and Next Steps
Requirements
Target Audience
AI engineers and architects designing agentic systems for production environments. Cybersecurity, risk, and compliance specialists managing AI assurance in regulated sectors such as finance, healthcare, and consulting. Senior developers and solution leads integrating multi-modal and multi-agent functionalities into enterprise platforms.
Testimonials (3)
The trainer is patient and very helpful. He knows the topic well.
CLIFFORD TABARES - Universal Leaf Philippines, Inc.
Course - Agentic AI for Business Automation: Use Cases & Integration
Good mixvof knowledge and practice
Ion Mironescu - Facultatea S.A.I.A.P.M.
Course - Agentic AI for Enterprise Applications
The mix of theory and practice and of high level and low level perspectives
