Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
Overview of LLM Architecture and Attack Surface
- The construction, deployment, and API access methods for LLMs
- Essential components of LLM application stacks, including prompts, agents, memory, and APIs
- Common sources of security issues in real-world deployments
Prompt Injection and Jailbreak Attacks
- The nature of prompt injection and associated risks
- Scenarios involving direct and indirect prompt injection
- Techniques used to bypass safety filters (jailbreaking)
- Strategies for detection and mitigation
Data Leakage and Privacy Risks
- Preventing accidental data exposure through model responses
- Addressing PII leaks and improper model memory usage
- Designing privacy-focused prompts and Retrieval-Augmented Generation (RAG) systems
LLM Output Filtering and Guarding
- Employing Guardrails AI for effective content filtering and validation
- Establishing clear output schemas and constraints
- Monitoring and logging unsafe or inappropriate outputs
Human-in-the-Loop and Workflow Approaches
- Optimal points for introducing human oversight
- Managing approval queues, scoring thresholds, and fallback processes
- Calibrating trust levels and leveraging explainability
Secure LLM App Design Patterns
- Applying least privilege and sandboxing to API calls and agents
- Implementing rate limiting, throttling, and abuse detection
- Ensuring robust chaining with LangChain and effective prompt isolation
Compliance, Logging, and Governance
- Guaranteeing the auditability of LLM outputs
- Maintaining traceability and version control for prompts
- Aligning systems with internal security policies and regulatory requirements
Summary and Next Steps
Requirements
- A solid grasp of large language models and prompt-based interfaces
- Practical experience developing LLM applications in Python
- Knowledge of API integrations and cloud-based deployment strategies
Target Audience
- AI developers
- Application and solution architects
- Technical product managers leveraging LLM tools