Prompts and context
Prompt injection, indirect instructions, system-prompt leakage and unsafe context boundaries.
AI security engineering for production systems
AI security consulting for LLMs, RAG and agents: Bettag Systems reviews your architecture, tests attack paths and helps implement fixes. Franz Bettag works directly with your engineering and operations teams in the Nuremberg region and across Germany. The goal is verifiable controls and workflows your team can maintain.
What gets secured
RAG and data
Poisoning, manipulated sources, authorization failures, data exfiltration and missing provenance.
Tools and agents
Over-privileged tools, unsafe action chains, missing approvals and uncontrolled side effects.
Operations
Evals, audit logs, monitoring, cost controls, incident response and secure rollouts.
How the work runs
1. Understand the system
Map architecture, data flows, models, retrieval, tools, permissions and human approval points.
2. Test attack paths
Build a threat model and reproducible tests for injection, exfiltration, poisoning and tool misuse.
3. Remove risk
Implement controls or hand over a prioritized backlog with clear ownership.
4. Secure operations
Anchor regression evals, logging, alerts, approvals and runbooks in production.
Deliverables
Threat model
Concrete assets, trust boundaries, attacker goals and traceable attack paths.
Security findings
Reproducible findings with risk, evidence and actionable remediation.
Secure production path
Prioritized architecture and implementation work, including evals and operational controls.
Continuing education and practical work
Course certificates for Claude, MCP and agents
Franz Bettag’s Anthropic course certificates complement his hands-on work on AI systems. The certificate page shows the original documents; the bug bounty case study explains model roles, tools and verification in his own system.
From this work to your project
Which review fits your system?
An LLM security audit examines your AI application, its data access and tools. An external penetration test examines the authorized attack surface of your websites, APIs and services. Platform engineering provides the operational foundation for both. We agree scope and implementation with the teams responsible.
