Senior Applied AI Engineer, Quizlet

• 5-8+ years of industry experience in applied ML/AI or ML‑heavy software engineering
• BS/MS in CS, ML, or related quantitative field (or equivalent experience)
• Experience building ranking/personalization or search systems (retrieval, Two‑Tower/dual encoders, multi‑task rankers) and contributing to online metric improvements (e.g., CTR, session depth, retention)
• Hands‑on experience with LLM/GenAI systems: data curation, fine‑tuning (SFT/PEFT, preference optimization), prompt engineering, evaluation, and productionization considerations (latency/cost/safety)
• Strong skills in Python/PyTorch, data and feature engineering, distributed training/inference on GPUs, and familiarity with modern MLOps (model registry, feature stores, monitoring, drift)
• Solid experiment design (offline/online), metrics literacy, and ability to translate product goals into modeling solutions
• Strong collaboration skills and eagerness to learn from senior engineers; some experience mentoring junior teammates is a plus

Desirable

• As an Applied AI Engineer, you will be working at the forefront of our AI strategy, shaping Quizlet’s AI develop in one of the two complementary domains:
• Personalization & Ranking – retrieval and ranking systems that match learners with the right content, experiences, and monetization moments across surfaces (search, feed, notifications, ads)
• Generative AI & Agentic Systems – LLM‑powered tutoring, content understanding/synthesis, and tools that boost learner outcomes and creator productivity
• You will work on a variety of models and modeling systems (from Two‑Tower retrieval and multi‑task rankers to RAG/LLM pipelines), ensure robust evaluation and responsible deployment
• Contribute to the technical roadmap for applied AI across personalization, ranking, search, recommendations, and GenAI/LLM systems; help connect modeling work to business metrics (engaged learners, conversion, retention, revenue)
• Build components of end‑to‑end ML systems: candidate sourcing, embedding platforms & ANN retrieval, multi‑stage ranking (early/late), and value modeling with guardrails for fairness and integrity
• Implement LLM‑based features: build RAG pipelines, apply instruction‑/preference‑tuning techniques (e.g., SFT/DPO), optimize prompts, and improve latency/cost‑aware inference; contribute to offline evals + human‑in‑the‑loop and online success metrics
• Help develop "Learner 360" representations by working with behavior signals, explicit inputs, and conversational context to create robust embeddings reused across surfaces
• Support evaluation infrastructure: contribute to the eval harness for both ranking and generative systems (offline metrics like NDCG/AUC/BLEU/BERTScore; quality/safety scorecards), and help close the loop with online A/B experiments
• Ship reliable systems at scale: ensure training‑serving consistency, implement drift detection, follow canarying/rollback protocols, participate in on‑call rotation for model services, and maintain strong CI/CD for features & models
• Collaborate with and learn from senior ML/SWE teammates; write high‑quality code and follow best practices for experimentation rigor and reproducibility
• Work closely with Product, Design, Legal, and Data Science on objectives, tradeoffs, and responsible AI practices
• Stay current with ML research (RecSys, LLMs, multimodal) and propose new methods that could improve learner outcomes