Staff Machine Learning Engineer, Ads Foundational Representations

Reddit is a community of communities. It’s built on shared interests, passion, and trust, and is home to the most open and authentic conversations on the internet. Every day, Reddit users submit, vote, and comment on the topics they care most about. With 100,000+ active communities and approximately 130 million daily active unique visitors, Reddit is one of the internet’s largest sources of information. For more information, visit www.redditinc.com.

Location: Reddit has a flexible first workforce. Don't live near our office? No worries: you can work remotely from anywhere in the UK or the Netherlands.

The Ads Foundational Representations (AFR) team develops signals and representations of Reddit’s core entities (ads, posts, users, and so on), capturing the semantic, contextual, and behavioral information that Reddit Ads needs. We work on building embeddings to understand content and users' interests based on the content they engage with.

Our team has the potential to highlight one of Reddit's biggest differentiators: genuinely curated, high-quality, extremely relevant, and daily updated organic content. We are a Machine Learning/Data heavy team with a focus on the following areas:

• Multimodal & Content Embeddings - Make sense of organic (posts, comments, subreddits) and promoted (ads, shopping products, their landing pages) text and media content by embedding them into a shared space.
• Contextual and Behavioral Relevance - Working with Product & Data Science, establishing definitions of what ads are relevant to users and the content we show them next to, building metrics and fine-tuning embeddings to better reflect relevance.
• Knowledge Graph Embeddings - Building representations for the Knowledge graph entities, e.g., intellectual properties/brands, to be used for high-precision targeting & business insights.
• User Intent Modeling - Leveraging various techniques to introduce user representations based on the content they interact with: batch & real-time sequence modeling, LLM summarization, etc.
• LLM-based Representati