Responsibilities
- Define the long-term technical vision and roadmap for user modeling and generative personalization, pinpointing core machine learning capabilities that can be leveraged across search, recommendations, conversational AI, and other personalized services.
- Design and build large-scale systems that infer user preferences, real-time intent, behavioral trends, interests, and contextual signals from diverse interaction data across platforms.
- Create reusable user embeddings and foundational models that serve multiple downstream applications such as retrieval, ranking, recommendation, personalized conversations, targeting, and user engagement.
- Pioneer generative recommendation techniques by investigating foundation models, generative retrieval methods, sequence modeling, and unified representations to enhance traditional retrieval and ranking systems.
- Lead platformization efforts by transforming proven modeling techniques into shared assets—including representations, models, features, APIs, and serving infrastructure—used across multiple teams.
- Develop robust evaluation frameworks, both offline and online, to assess the performance, generalization, and incremental value of user understanding and generative models in real-world applications.
- Collaborate with senior machine learning and platform engineering leaders to design production-grade architectures for large-scale training, real-time inference, and low-latency model serving.
- Monitor advancements in generative recommendation, foundation models, user modeling, generative retrieval, and representation learning, and drive promising research prototypes into scalable production solutions.
Work Arrangement
On-site — Grab One North Singapore office
