Clera Research Engineer, Privacy and Anonymization at Clera building privacy systems for AI infrastructure to protect sensitive data for model training and evaluation. Develop end-to-end methods for data anonymization and privacy risk evaluation.
Responsibilities
Build privacy and anonymization systems for an AI infrastructure team, helping make sensitive real-world data safe and useful for model training and evaluation. You will develop end-to-end methods that protect sensitive information while preserving the structure and signal needed by downstream workflows.
Build systems to detect PII, quasi-identifiers, credentials, and other sensitive information, and apply transformations suited to each data type and use case.
Develop and benchmark detection approaches combining rules, statistical models, classifiers, and LLM-based methods.
Create production pipelines that anonymize raw data before processing, training, evaluation, or synthetic data generation.
Develop evaluation frameworks for privacy risk and retained utility, including recall-weighted metrics, leakage tests, and adversarial re-identification attempts.
Design robust systems that handle new data sources, schema drift, unusual formats, and sensitive information in unexpected fields.
Work with engineering, research, operations, and customers to turn privacy requirements into practical safeguards.
Qualification
Experience with information extractionKnowledge of redactionExperience handling schema drift
Required
At least 2 years of experience building production data or ML systems in Python, with strong proficiency in the language.
A background in a technical field and hands-on experience detecting, removing, or transforming PII while preserving useful data characteristics.
Experience with information extraction, named-entity recognition, classification, or related approaches for identifying rare or sensitive content.
Experience building end-to-end data processing pipelines and benchmarking methods across recall, precision, latency, cost, and data utility.
Knowledge of redaction, masking, pseudonymization, anonymization, and synthetic data generation.
Experience handling schema drift, unusual data formats, and edge cases; familiarity with privacy-enhancing techniques such as differential privacy, k-anonymity, secure aggregation, or format-preserving encryption is valuable.
Experience with low-latency or high-throughput ML systems and sensitive data in areas such as healthcare, finance, or security is beneficial.
Preferred
Familiarity with privacy-enhancing techniques such as differential privacy, k-anonymity, secure aggregation, or format-preserving encryption
Experience with low-latency or high-throughput ML systems and sensitive data in healthcare, finance, or security