Applied Scientist II — NLP & Multimodal Search

Amazon Science; Seattle (WA); On-site; USD 143,000 - 193,000; Full time

Job summary

Amazon Science in Seattle is seeking an Applied Scientist II to advance Natural Language Search, interpreting complex shopping queries and ranking products using text and images. You will build scalable relevance models spanning query understanding and multimodal matching at Amazon scale.

You will ship end-to-end ranking systems, craft data and evaluation methods, and collaborate with teams to move models from prototype to production.

Qualifications

• PhD or Master's degree with 2+ years in CS, CE, ML or related field.
• Experience programming in Java, C++, Python or related language.
• Experience in ML, RL, DL, CV, NLP or related applications.

Responsibilities

• Design, train, and ship deep-learning ranking and semantic-matching models over text and product images to improve search relevance.
• Build training data and evaluation methods: synthetic/historical labels, hard-negative mining, targeted sampling.
• Train and adapt vision-language encoders so product images become a core relevance signal and verify visual constraints that text misses.
• Build joint text-image embeddings for retrieval and ranking with contrastive training and modality fusion.
• Develop signals that match product attributes to customer queries.
• Run offline and online A/B experiments, analyze precision/recall, and iterate to launch.

Skills

PhD or MS+2y

Java/C++/Python

ML/NLP/DL/CV

Job description

Amazon Science in Seattle is seeking an Applied Scientist II to advance Natural Language Search, interpreting complex shopping queries and ranking products using text and images. You will build scalable relevance models spanning query understanding and multimodal matching at Amazon scale.

You will ship end-to-end ranking systems, craft data and evaluation methods, and collaborate with teams to move models from prototype to production.