岗位介绍：作为美图用户画像的核心算法角色，你将参与全域用户画像体系的建设与迭代，基于海量多模态用户行为完成用户偏好、兴趣意图建模，同时结合大模型与 Agent 技术探索新一代用户理解方案，画像能力深度赋能 AIGC 创作、影像产品及搜索推荐业务。你将深度参与算法方案设计、技术落地与业务闭环，通过 A/B 实验持续驱动用户体验与业务收益提升。我们寻找兼具扎实算法工程能力与业务洞察力的实践者，期待你把用户画像、大模型 Agent、搜索推荐的技术积累，落地到美图影像 AI 产品，打造面向海量用户的智能个性化能力。

岗位职责

1、负责美图全域用户画像体系建设，优化数据处理与标签生产流程，保障画像数据质量，支撑影像产品、AIGC、个性化推荐业务；

2、构建迭代用户兴趣、意图、分层等画像模型，基于多源行为数据挖掘用户偏好，产出标签与用户向量；

3、结合大模型、Agent技术探索新一代用户理解方案，赋能AI Agent个性化能力；画像能力对接产品运营策略、搜索推荐链路等，参与A/B实验，驱动业务指标提升；

4、对接产品、运营，输出数据洞察，迭代算法方案，持续优化系统性能。

任职要求

1、计算机/统计/AI相关硕士及以上，3年+算法开发经验，具备扎实的数学和数据挖掘功底，熟悉特征工程、用户建模，有大规模用户画像落地经验的优先；

2、掌握Python/Java/Scala至少一门语言，至少掌握PyTorch/LightGBM/XGBoost一个机器学习工具，熟悉Hive、Spark、Flink等大数据组件的优先；

3、了解大模型、Agent相关技术，能够落地大模型增强用户理解相关方案；

4、对数据敏感，具备业务理解能力，善于通过实验验证算法效果，良好沟通与团队协作。

优先

1、有搜索推荐算法实战经验优先

2、有工具/影像/AIGC产品画像经验优先

Algorithm Engineer – User Profiling · Xiamen

At Meitu, user profiling is not a support layer — it is the foundation underneath products like BeautyPlus, AirBrush, and our AIGC creation flows. We are building a global user understanding system that turns multimodal behavior into reliable preference, interest, and intent signals. You will work across the full loop: data pipeline, model design, offline evaluation, online A/B testing, and iteration. This role sits at the intersection of classical user modeling and newer LLM/Agent-based understanding, with real room to explore and ship.

Your Core Work:

· Own the global user profiling data pipeline and tag production process, including quality standards that support imaging products, AIGC features, and personalized recommendation;

· Design and iterate user interest, intent, and segmentation models, turning multi-source behavioral data into user tags and embeddings;

· Explore LLM and Agent-based approaches for next-generation user understanding, and connect those signals to AI Agent personalization;

· Partner with product, operations, search, and recommendation teams to run A/B experiments and translate profiling output into measurable business impact;

· Use data insights to refine algorithm design and improve system performance end to end.

What We Hope You Have:

· MS or higher in Computer Science, Statistics, AI, or a related field, with 3+ years of algorithm development experience;

· Solid grounding in mathematics and data mining, with hands-on experience in feature engineering and user modeling;

· Proficiency in at least one of Python, Java, or Scala, and practical experience with at least one ML framework such as PyTorch, LightGBM, or XGBoost;

· Working knowledge of LLM and Agent technologies, with the ability to apply them to user understanding problems;

· Strong data intuition and business sense, with experience validating algorithm changes through experiments and collaborating across teams.

Nice to Have:

· Production experience building large-scale user profiling systems;

· Familiarity with big data components such as Hive, Spark, or Flink;

· Hands-on experience with search or recommendation algorithms;

· Background working on profiling for utility, imaging, or AIGC products.