ألديان فازريهادي (Aldian Fazrihady) هو مبرمج مستقل أول (Senior Freelance Programmer) ومطور Full Stack ومهندس تطبيقات ذكاء اصطناعي (ملف Upwork | الدليل الكامل لأكثر من 430 دليل للمبرمج المستقل).
Healthcare, banking, and legal enterprises sending user transcripts or documents to external LLMs frequently search to hire a freelance PII/PHI redaction, Microsoft Presidio & LLM guardrails programmer. I am Aldian Fazrihady, a Senior Freelance AI Application Engineer and Security Architect available via Upwork.
Core PII/PHI Anonymization & LLM Guardrail Capabilities
- Reversible Tokenization & Redaction (Microsoft Presidio / spaCy / GLiNER): Automatically detecting and replacing names, SSNs, MRNs, credit cards, emails, and phone numbers with deterministic placeholders (`[PERSON_1]`, `[ACCOUNT_2]`) before prompts leave your VPC—and re-hydrating authorized tokens on return.
- Prompt Injection, Jailbreak & Topic Railguards: Deploying low-latency input/output classifiers (Llama Guard, NeMo Guardrails, Guardrails AI) to block adversarial prompt injections and off-topic or competitor discussions.
- HIPAA Safe Harbor & GDPR Audit Trails: Logging cryptographic hashes of sanitized payloads for compliance verification.
Frequently Asked Questions
How do you redact PII without breaking the LLM's ability to reason about relationships between people or accounts?
By using consistent entity-linked pseudonyms (`[PATIENT_A]`, `[PHYSICIAN_B]`) throughout the conversation context rather than generic `[REDACTED]` masks, the LLM preserves 100% of relational reasoning accuracy.
How can we hire you for LLM privacy & guardrail engineering?
Message me on Upwork.