Federated/DP Privacy ML Scientist

HushOne, Inc.; Kirkland (WA); Hybrid; USD 120,000 - 180,000; Full time

Benefits offered by this job

Stock options

401(k) with company match

Gym membership

Remote-friendly around family

Annual performance bonus

Job summary

HushOne, Inc. seeks a privacy-focused ML researcher to evaluate federated schemes, model adversaries, and quantify privacy-utility tradeoffs. You will collaborate with cryptography and product teams to deploy guarantees and clear consent.

The role emphasizes rigor in privacy or ML research, explaining why on-device data is not inherently private, and defining adversaries before guarantees.

Qualifications

• Rigorous privacy or machine-learning research expertise.
• Explain why on-device data is not automatically private, and design accordingly.
• You define an adversary before claiming a guarantee.
• You can quantify a privacy and utility tradeoff honestly.

Responsibilities

• Evaluate a proposed federated training scheme for update leakage.
• Assess threats from malicious clients and withdrawal after participation.
• Collaborate with product teams to deploy privacy guarantees and understandable consent.

Skills

Privacy research

ML research

Adversary modeling

Privacy-utility tradeoffs

Job description

HushOne, Inc. seeks a privacy-focused ML researcher to evaluate federated schemes, model adversaries, and quantify privacy-utility tradeoffs. You will collaborate with cryptography and product teams to deploy guarantees and clear consent.

The role emphasizes rigor in privacy or ML research, explaining why on-device data is not inherently private, and defining adversaries before guarantees.