岗位职责
您将不仅仅是构建模型的工程师，您需要深入理解用户和产品，将业务问题转化为算法问题，并利用先进的机器学习技术，实现业务指标的提升。 主要职责：
- 业务洞察与问题定义：深入理解业务场景，与产品、运营等团队紧密协作，挖掘用户在推荐、排序、偏好预测等方面的核心痛点与机会点。负责将模糊的业务需求转化为清晰、可量化、可解决的算法问题。
- 算法模型开发与优化：设计、实现并优化模型算法和策略，包括但不限于经典机器学习模型、Embedding、序列模型、深度学习模型、多目标建模等。
- 系统落地与效果评估：推动算法模型在线上系统部署、AB测试和效果监控；设计和分析AB实验，科学评估算法迭代对业务指标的影响，并能清晰向非技术团队解读数据结果；持续监控线上模型表现，快速定位并修复模型异常问题。
职位要求
- 数学、统计、计算机等相关专业硕士及以上学历；
- 熟练使用至少一种主流机器学习框架（如Tensorflow,Pytorch,Scikit-learn)，精通Python编程语言;熟练掌握SQL;
- 熟悉业界常用的机器学习算法、推荐排序算法和技术。熟悉黑盒模型的可解释性方法;
- 良好的逻辑思维能力，对数据敏感，能够发现关键数据、抓住核心问题，学习能力强，有强烈的责任心，良好的沟通协调能力；
- 具备强烈的业务意识和数据驱动思维，能独立完成从业务理解到模型上线的完整闭环。 Job Description You will be more than just an engineer building models — you'll need to deeply understand users and products, translate business problems into algorithmic challenges, and leverage advanced machine learning techniques to drive business metric improvements. Key Responsibilities: Business Insight & Problem Definition: Develop a deep understanding of business scenarios and work closely with product, operations, and other teams to identify core pain points and opportunities in recommendation, ranking, preference prediction, and related areas. Translate ambiguous business requirements into clear, quantifiable, and solvable algorithmic problems. Algorithm & Model Development and Optimization: Design, implement, and optimize model algorithms and strategies, including but not limited to classical machine learning models, embeddings, sequence models, deep learning models, and multi-objective modeling. System Deployment & Performance Evaluation: Drive the deployment of algorithmic models in production systems, A/B testing, and performance monitoring; design and analyze A/B experiments to scientifically evaluate the impact of algorithm iterations on business metrics, and clearly communicate data-driven insights to non-technical teams; continuously monitor online model performance, rapidly diagnose and resolve model anomalies. Qualifications: Master's degree or above in Mathematics, Statistics, Computer Science, or a related field. Proficient in at least one mainstream machine learning framework (e.g., TensorFlow, PyTorch, Scikit-learn); expert-level Python programming skills; strong proficiency in SQL. Familiar with industry-standard machine learning algorithms, recommendation and ranking algorithms, and related technologies. Familiar with interpretability methods for black-box models. Strong logical thinking skills and data sensitivity — able to identify key data signals and pinpoint core problems. Fast learner with a strong sense of ownership and excellent communication and collaboration skills. Strong business acumen and data-driven mindset; capable of independently driving the full lifecycle from business understanding to model deployment in production.
