Privacy Preserving Cryptography Questions

Techniques that combine cryptography and privacy engineering to enable secure computation and data protection. Core topics include homomorphic encryption for computing over encrypted data, secure multi party computation for collaborative computation without revealing inputs, differential privacy methods for statistical analysis with privacy guarantees, oblivious transfer and related secure protocol primitives, and zero knowledge proof systems for proving statements without revealing secrets. Coverage includes practical use cases, performance and scalability limitations, parameter and threat model selection, trade offs between privacy and utility, deployment challenges, and when to prefer one approach over another.

MediumSystem Design

83 practiced

Design a monitoring and governance system to track privacy-budget (ε, δ) consumption across multiple ML services and teams in production. Include APIs or SDK calls for services to log privacy-consuming operations, a central privacy ledger, dashboards, alert rules for budget exhaustion, enforcement mechanisms (rate-limiting or automatic throttling), and audit trails for compliance. Discuss trust boundaries and how to prevent teams from bypassing the ledger.

HardTechnical

83 practiced

Explain how security proofs for MPC differ between semi-honest and malicious models. Describe generic techniques for compiling a semi-honest protocol into a maliciously secure one (e.g., authenticated shares, cut-and-choose, zero-knowledge proofs, commit-and-open), and estimate the practical performance overhead and engineering challenges of such transformations in large-scale ML workloads.

HardTechnical

74 practiced

Using Microsoft SEAL (or its Python bindings), implement an inference pipeline for linear regression using CKKS: show code snippets for key generation, encoder/decoder setup, encoding of a float feature vector, encryption, server-side homomorphic linear evaluation (weighted sum), decryption, and decoding. Discuss how scale and modulus selection affects precision and noise budget and what test you would run to validate numeric correctness.

EasyTechnical

76 practiced

Describe secure multi-party computation (MPC) at a high level. Explain the difference between the honest-but-curious (semi-honest) model and the malicious adversary model. Give two practical ML use-cases where MPC is preferable to HE or DP, and mention mature MPC frameworks an ML engineer might evaluate (e.g., MP-SPDZ, Sharemind, CrypTen).

HardTechnical

92 practiced

You're optimizing HE-based inference for a relatively large model. Provide a concrete optimization plan: how to use ciphertext packing (slots) efficiently, choose CKKS polynomial modulus degree and coefficient modulus to balance security and noise budget, approximate activation functions with low-degree polynomials, apply quantization and pruning, and leverage hardware (SIMD/GPU) or multi-threading. Give rough order-of-magnitude expectations for latency improvements from each technique.

Unlock Full Question Bank

Get access to hundreds of Privacy Preserving Cryptography interview questions and detailed answers.

Join thousands of developers preparing for their dream job.