✦ Luna Orbit — AI & Machine Learning

Sr. Applied Scientist, Ads AI Core Infrastructure

at Amazon.com

📍 US, NY, New York Hybrid Posted April 01, 2026
Type Full-Time
Experience senior
Exp. Years 3+ years
Education PhD, or Master's degree
Category AI & Machine Learning

Sr. Applied Scientist building AI infrastructure for ads, researching agent-data interaction, RAG embeddings, and MCP server infrastructure, balancing research and production.

  • Agent orchestration & optimization research
  • Large Language Model context & token optimization
  • RAG-based embeddings & semantic search
  • Experimentation & productionization
  • Cross-functional collaboration & thought leadership
  • Research publication & innovation

Research and productionize ML approaches for agent-data interaction, including MCP servers, CodeAct, RAG embeddings, and LLM optimization for real-time advertiser data.

The ideal candidate is a senior applied scientist with 3+ years of ML experience, strong in multi-language programming (Java/C++/Python), expert in RAG embeddings and real-time data, and capable of productionizing research with ML-enabled advertising agents.

3+ years building machine learning models for business application experiencePhDor Master's degree and 6+ years of applied research experienceExperience programming in JavaC++PythonExperience with neural deep learning methods and machine learning
Experience with modeling tools such as Rscikit-learnSpark MLLibMxNetTensorflownumpyscipyExperience with large scale distributed systems such as HadoopSpark
JavaC++PythonRCodeActModel Context ProtocolMCP serversCodeAct patternsTensorFlowSparkHadoopMxNetscikit-learnNumpySciPy
3+ years building ML models for business; JavaC++Python; neural deep learning; CodeAct; RAG-based embeddings; Model Context Protocol; MCP servers; large language models
JavaC++Pythonneural deep learning methodsRAG-based embeddingsCodeActModel Context ProtocolMCP serverslarge language modelsCodeAct patternssemantic searchreal-time data deliveryadvertiser dataproductionization
collaborationmentoringcommunicationleadershipproblem-solving
Industry SaaS
Job Function Research and develop novel ML approaches for agent-data interaction and productionize innovations at Amazon.
Role Subtype AI Research Scientist
Tech Domains Python, Java, C++, TensorFlow, Spark, Hadoop, Model Context Protocol, MCP servers, CodeAct, RAG-based embeddings, large language models, scikit-learn, MxNet, Numpy, SciPy
Sr. Applied ScientistAds AI Core InfrastructureCodeActRAG-based embeddingsModel Context ProtocolMCP serverslarge language modelsagent orchestrationcontext optimizationsemantic searchreal-time advertiser dataproductionizationadvertiser dataCodeAct patternsTensorFlowSparkHadoopPythonJavaC++neural deep learning
Apply for this Position →

Get matched to jobs like this

Luna finds roles that fit your skills and career goals — no endless scrolling required.

Create a Free Profile