OpenAI-compatible reverse proxy: inbound threat detection, reversible DLP, routing, and evals. MSc thesis deposited (pending defense).
Drop-in middleware between apps and any LLM provider. Pipeline: sanitisation and OWASP-style filters, LLM-as-judge in parallel with DistilBERT, bilingual DLP, then LiteLLM routing. Evaluation battery 17/17 PASS; ~175 pytest. Repo: github.com/r4fik1/ai-security-gateway.
STRIDE for LLM-enabled features: treat the model as an untrusted component.
A diagram, trust boundaries, and a human gate — the same habit as a Vault ceremony, applied to tools and mitigations.
Thesis note: baseline versus protected profile, 17/17 PASS, and what that number does not prove.
A replayable suite — layer hit, HTTP status, latency — instead of a screenshot of a chat.
Security as middleware: an OpenAI-compatible reverse proxy for LLM traffic.
If every client talks to the same Base URL, notebooks, chatbots and batch jobs share one pipeline.
DevSecOps Engineer · Swisscom
Relocating to Switzerland · CAISP · MSc AI (thesis deposited)
I turn security process into platforms. At Swisscom I run DevSecOps for 1000+ developers and high-assurance HashiCorp Vault operations, and I ship security-transparency products so leadership sees posture without spreadsheet archaeology. Thesis: an OpenAI-compatible AI Security Gateway.