Secure LLM inference using AWS Nitro TPM attestation with Zero Operator Access. Deploy models in end user's accounts while protecting model weights. - View it on GitHub