Member of Technical Staff (AI Engineer)

About this role

About the role: - You will tackle technical challenges across synthetic data generation, LLM outputs, privacy protection, and large-scale data systems as a hands-on engineer - This is an individual contributor position focused on solving problems through code and system design, not management - You'll partner with researchers, engineers, domain specialists, and customers to shape how data flows through production systems What you'll work on: - Building and maintaining pipelines that generate synthetic or LLM-generated data at scale for model training - Designing systems to de-identify, anonymize, or transform sensitive data while preserving relationships and structural integrity - Creating benchmarks, evaluation frameworks, and annotation systems for LLMs and AI agents - Assessing data quality through metrics, hypothesis testing, and root-cause analysis when things go wrong - Weighing trade-offs between quality, scale, reliability, and whether systems can be reused across projects - Contributing to post-training workflows, model evaluation pipelines, and validation methodologies What you'll need: - 2–10 years of professional experience - Demonstrated, recent hands-on work in at least one of: synthetic data, LLM-generated data, post-training and evaluation, privacy engineering, or large-scale anonymization and de-identification - Proficiency in Python with the ability to write clean, efficient, and scalable code for complex, large datasets - Solid judgment about data quality, risk assessment, and utility; comfort with metrics, statistical testing, and debugging failures - Skill in designing systems and evaluating where to compromise on quality, performance, scale, reliability, and reusability - Comfort working with ownership in environments that change quickly and where requirements are not fully defined upfront Especially valuable: - Experience running large-scale pipelines that generate synthetic or LLM-created training data - Background building or operating data de-identification or anonymization systems, especially those that maintain graph or relational structure - Work on LLM or agent evaluation, assessment methodologies, data-quality frameworks, or annotation infrastructure - Research or applied experience in reinforcement learning, model alignment, behavior analysis, human-in-the-loop systems, or synthetic data - Exposure to regulated or sensitive data environments, re-identification risk modeling, or privacy audits - Published research, substantial open-source work, or technical leadership in machine learning systems, data engineering, or AI research - Track record of turning research into production-ready, reusable systems or scaling customer solutions into platforms