出海雷达

高级数据科学家(Look-2-Book)

Senior Data Scientist (Look-2-Book)

开发工程限定地区(需当地身份)日间重叠仅 1 小时,需熬夜配合
公司Emerging Travel Group
薪资未公开
工作地点Portugal
地域资格限定地区(需当地身份)
时区要求日间重叠仅 1 小时,需熬夜配合
用工类型Full Time
发布时间14 天前
数据来源Himalayas
前往 Himalayas 查看并投递 →
注意地域限制:该职位明确限定在 Portugal 招聘。如果你是位于中国大陆的求职者,通常需要当地工作身份才能投递,或需与雇主确认是否接受独立合同(Contractor)形式合作。
作息提示:日间重叠仅 1 小时,需熬夜配合。
  • 端到端机器学习负责:从提出假设和任务设定到实现可衡量的业务影响,主导机器学习项目。您将负责整个流程,确保快速上市时间,并在生产环境中保持模型质量。
  • 跨职能协作:与数据工程师紧密合作,构建稳健的生产数据流水线;与数据分析师合作,深入业务背景,定义指标,并设计和评估A/B测试。
  • 领域特定问题解决:将复杂的旅行/B2B业务问题转化为数学和机器学习建模任务。主要挑战包括:Look-to-Book优化:设计和部署评分模型,准确预测搜索查询转化为预订(look-to-book)的概率,模拟该可能性如何随价格和可用性动态变化,并优化整个搜索到预订流程;多目标优化:同时优化利润率/盈利能力和操作事件最小化;考虑B2B客户偏好的复杂权衡建议;缓存新鲜度预测:预测价格/优惠在预订前过时的可能性,帮助在数据准确性、速度和系统负载之间找到合适的平衡。
  • 模型开发:开发、训练和验证机器学习模型,以测试产品假设并主动提升B2B用户体验。

要求
必须具备:

  • 经验:至少4年以上作为数据科学家或机器学习工程师的实际经验。
  • 商业思维:能够深入数据分析,识别根本业务问题,并将业务目标转化为明确的机器学习任务,而不仅仅是孤立地调整指标。
  • 机器学习堆栈与算法:对经典机器学习(如梯度提升如CatBoost/LightGBM、分类、回归)有扎实的理解。
  • 匹配、推荐系统与排序:在实体匹配任务上有实际经验,并对推荐系统和排序方法有深入理解(KNN、FAISS、Learning-to-Rank,点对点/成对/列表方法)。
  • 大数据堆栈:有处理海量数据的经验(我们每天处理TB级数据)。优秀的SQL技能,并有使用PySpark构建流水线的实际经验。
  • 工程独立性:能编写高质量的Python代码,具备编写测试的能力,并准备好将模型部署到生产环境(有Airflow和Python微服务的经验)。

加分项:

  • 行业经验:有旅行行业经验
查看英文原文
  • End-to-End ML Ownership: Lead ML projects from formulating hypotheses and task setting to delivering measurable business impact. You will be responsible for the entire cycle, ensuring fast time-to-market and maintaining model quality in production.
  • Cross-functional Collaboration: Work closely with Data Engineers to build robust production data pipelines, and with Data Analysts to dive into business context, define metrics, and design/evaluate A/B tests.
  • Domain-Specific Problem Solving: Translate complex travel/B2B business problems into mathematical and ML modeling tasks. Key challenges include: Look-to-Book Optimization: Design and deploy scoring models to accurately predict the probability of a search query converting into a booking (look-to-book), model how this likelihood changes dynamically with price and availability, and optimize the entire search-to-book funnel / Multi-objective Optimization: Simultaneous optimization of margin/profitability and operational incident minimization; sophisticated trade-off recommendations factoring in B2B client preferences / Cache Freshness Prediction: Predict the likelihood of rates/offers becoming stale before booking, helping find the right balance between data accuracy, speed, and system load.
  • Model Development: Develop, train, and validate machine learning models to test product hypotheses and proactively improve the B2B user experience.

Requirements
Must Have:

  • Experience: At least 4+ years of hands-on experience as a Data Scientist or ML Engineer.
  • Business Mindset: Ability to dig into data, identify root business problems, and translate business objectives into clear ML tasks rather than just tuning metrics in a vacuum.
  • ML Stack & Algorithms: Solid knowledge of classic Machine Learning (Gradient Boosting like CatBoost/LightGBM, Classification, Regression).
  • Matching, RecSys & Ranking: Proven experience with entity matching tasks, and a deep understanding of Recommendation Systems and Ranking approaches (KNN, FAISS, Learning-to-Rank, pointwise / pairwise / listwise approaches).
  • Big Data Stack: Confident experience working with massive datasets (we process terabytes daily). Excellent SQL skills and practical experience building pipelines with PySpark.
  • Engineering Independence: Production-quality Python code, ability to write tests, and readiness to bring models to production (experience with Airflow and Python microservices).

Nice to Have:

  • Industry Experience: Experience in TravelTech, E-commerce, or B2B APIs (understanding of hotels, rates, distributors).
  • NLP & Embeddings: Basic NLP skills, understanding the principles of text embeddings and how to train them (highly useful for matching textual rate descriptions, room types, and cancellation policies).
  • Optimization & Pricing: Experience with dynamic pricing, multi-objective optimization, or Next Best Action (NBA) systems.
  • Math & Stats: A solid foundation in probability theory and mathematical statistics for rigorous experiment design and model evaluation.

Benefits

  • A fully flexible work schedule — there’s no pressure to start work at exactly 9:00 AM; what matters is achieving results and moving forward;
  • Each person in our team is encouraged to choose their preferred work format. You can work fully remotely, come to the office, or choose a hybrid work model;
  • We are an ambitious and supportive team who love what they do, appreciate each other, and grow together;
  • The growth and development of each employee is our priority, so we have internal programs available for adaptation and training, development of soft skills and leadership abilities that are tailored individually to each employee;
  • We also provide partial compensation for employees participating in external training and conferences;
  • In tourism, it's difficult to grow without an excellent knowledge of English, and we support our employees' language learning goals — we organize group and individual lessons, plus speaking clubs with colleagues from all over the world;
  • And, of course, to encourage you to travel more, we offer corporate prices on hotels and other travel services;
  • We prioritize well-being and are committed to supporting the overall health and work-life balance at ETG. As part of this commitment, we provide MyTime Day Off - an extra day off that is designed to give our employees the flexibility to focus on important matters, whether it’s taking care of their health, mental recharge, addressing personal issues, or any other important activities.

Learn more about our data protection practices in our Privacy Policy:
Originally posted on Himalayas

本页面信息整理自 Himalayas,版权归原发布方所有。职位可能随时关闭,投递请以原始页面为准。 本站只做信息聚合展示,不参与招聘流程,也不向求职者收取任何费用。

该公司其他在招职位

高级会计师

Emerging Travel GroupKazakhstanTemporary15 天前
职能支持未标注地域

← 返回全部职位