Applied Scientist II, Alexa Daily Essentials Science and Analytics
Core
Design agentic AI systems that autonomously detect, diagnose, and act on signals across Alexa's product surface, while building evaluation science to validate performance.
Role type
Applied Scientist II (Agentic Systems & Evaluation)
Builds
Agentic AI systems, evaluation frameworks, and novel AI-powered features for Alexa
Domain
Consumer AI / Alexa / Agentic Systems
Deliverable
production ML models | product features
Required skills
Reinforcement Learning, Deep Learning, Natural Language Processing (NLP), Machine Learning, Java, C++, Python, RAG pipelines, automated evaluation frameworks
Preferred skills
Professional software development, novel algorithm design, agentic AI system design (tool-use, multi-step reasoning, multi-agent orchestration), LLM-as-judge systems
Technologies
Java, C++, Python, RAG, LLMs
Responsibilities
Design and implement agentic AI systems with multi-step reasoning and tool use; Apply reinforcement learning to improve agent routing and decision-making; Build evaluation and benchmarking frameworks including automated test suites and LLM-as-judge pipelines; Prototype and develop novel AI-powered features from concept to production; Build RAG pipelines and knowledge retrieval systems; Design and execute experiments including causal analysis and A/B tests
Seniority
Mid-Senior, hands-on IC