Staff Machine Learning Engineer, Ads Signals Intelligence & Information Retrieval

What you'll do

• Design and scale ML systems to extract semantic signals from structured and unstructured content.
• Develop retrieval and ranking pipelines using semantic embeddings and dense/sparse indexing.
• Fine-tune Large Language Models for content labeling, rewriting, and semantic similarity tasks.
• Construct knowledge graphs and entity linking systems to enrich creative and query signals.
• Build multimodal representations by combining text, image, and metadata signals.
• Develop content understanding components including entity extraction, topic modeling, and taxonomy generation.
• Own the experimentation, offline evaluation, and online validation of signal pipelines at scale.

What we're looking for

• Bachelor's in Computer Science, Machine Learning, Information Retrieval, NLP, or a related field.
• MS or PhD in Computer Science, Machine Learning, Information Retrieval, NLP, or a related field (preferred).
• 4+ years of experience in machine learning or applied research, focusing on retrieval, ranking, NLP, or content understanding.
• 7+ years of experience in machine learning or applied research, focusing on retrieval, ranking, NLP, or content understanding (preferred).
• Deep understanding of information retrieval, semantic search, and query-document matching.
• Strong hands-on experience with LLM fine-tuning, knowledge graph construction, and entity-centric modeling.
• Experience working with multimodal models, including text, vision, metadata, or audio-based representations.
• Proficiency in Python and experience with ML frameworks like PyTorch or TensorFlow.
• Background in statistical modeling, optimization, and ML theory.
• Exposure to ad tech domains such as auction modeling, targeting, attribution, or creative optimization (preferred).