Design and implement agent-assisted pipelines to parse, audit, and analyze massive-scale system logs. Define, instrument, and track system-level key performance indicators (KPIs), backend performance metrics, and product health indicators. Develop custom scripts, prompts, or agentic wrappers (e.g., in Python or R) to parse complex, messy infrastructure logs and transform them into digestible, statistically validated behavioral patterns.
Partner deeply with Infrastructure Engineers, Applied Data Scientists, and Machine Learning researchers to ensure our backend pipelines are correctly instrumented for downstream agentic analysis. Translate abstract system performance data (e.g., latency, GPU/TPU utilization, noise, storage toil) into quantifiable measures of user friction, efficiency losses, and "cognitive load".
Job description provided by Google.
View the original posting
