I implement access controls, monitoring, and spend limits for RAG, agents, and chatbots—so you avoid tool misuse, data exposure, and surprise bills.
What I Ship
- Access Controls & Guardrails: Input/output filtering, permission scoping, PII detection
- Monitoring & Alerting: Cost dashboards, anomaly detection, audit trails
- Spend Limits: Per-user budgets, rate limiting, token caps
Open Source
- LLM Production Safety Scanner — CLI tool for testing security, reliability, and cost controls in LLM apps
Links