At Quizlet, the business problem is to design and deliver AI-powered learning tools that scale across the world and unlock human potential, by making Quizlet feel uniquely tailored for every learner through cutting-edge machine learning, scalable infrastructure, and insights from learning science.
Requirements
- 12+ years of experience in applied machine learning or ML-heavy engineering, with deep expertise in personalization, ranking, or recommendation systems
- Proven ability to shape technical direction across multiple teams or disciplines, balancing long-term architectural vision with near-term product and business priorities
- Deep technical understanding of modern retrieval and ranking architectures (e.g., Two-Tower, deep cross networks, GNNs, MMoE, Transformers) and multi-stage RecSys pipelines
- Strong hands-on skills in Python and PyTorch, with expertise in data and feature engineering, distributed training and inference on GPUs, and familiarity with modern MLOps practices — including model registries, feature stores, monitoring, and drift detection
- Experience with large-scale embedding models and vector search systems (FAISS, ScaNN, or similar), including training, serving, and optimization at scale
- Expertise in experimentation and evaluation, connecting offline metrics (AUC, NDCG, calibration) with online A/B results to drive confident, data-informed decisions
- Commitment to collaboration and inclusion, fostering a culture that values diverse perspectives, constructive debate, and shared ownership of results
Responsibilities
- Work closely with other senior leaders to define and drive the long-term technical vision for personalization and recommendations across multiple Quizlet surfaces, ensuring alignment between modeling strategy, platform capabilities, and product roadmaps
- Architect and build large-scale personalization models across candidate retrieval, ranking, and post-ranking layers, leveraging user embeddings, contextual signals, and content features to power adaptive learning experiences
- Develop scalable retrieval and serving systems using modern architectures such as Two-Tower, deep ranking, and ANN-based vector search for real-time personalization at global scale
- Lead model training, evaluation, and deployment pipelines for retrieval and ranking systems, ensuring training-serving consistency, reliability, and robust monitoring
- Partner closely with Product and Data Science to translate learning objectives (e.g., engagement, retention, and mastery) into measurable modeling goals and experimentation frameworks
- Advance evaluation methodologies by refining offline metrics (e.g., NDCG, CTR, calibration) and online A/B testing to rigorously measure learner impact and model performance
- Collaborate with platform and infrastructure teams to optimize distributed training, inference latency, and cost-efficient serving in production environments
Other
- Minimum of three days per week in the office (Monday, Wednesday, and Thursday) and as needed by your manager or the company
- Exceptional communication and storytelling skills — able to distill complex technical concepts into clear narratives for executives, product partners, and non-technical audiences
- Demonstrated leadership through influence, guiding teams through ambiguity, aligning stakeholders around measurable goals, and ensuring accountability for impact
- Experience mentoring senior engineers and applied scientists, leading technical working groups, and driving cross-team innovation and standardization
- Commitment to equity, diversity, and inclusion, and a strong understanding of how to foster a culture of collaboration, inclusivity, and experimentation
