微软Applied Scientist 2
任职要求
Basic Qualifications: Master’s degree or above (or equivalent experience) in Computer Science, Engineering, Mathematics, Physics, or a related field. Strong programming skills with hands-on experience in managing large-scale data and machine learning pipelines. Deep understanding of open-source ML frameworks such as PyTorch, vLLM, and TensorRT-LLM (TRT-LLM). Solid knowledge of model optimization techniques, including quantization, pruning, and efficient inference. Preferred Qualifications: 1+ years of experience optimizing LLM inference using frameworks like vLLM or TRT-LLM. Practical experience in model compression and deployment within production systems. Experience designing agentic AI systems, such as multi-agent orchestration, tool usage, planning, and reasoning. Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, color, family or medical care leave, gender identity or expression, genetic information, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran status, race, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable laws, regulations and ordinances. If you need assistance and/or a reasonable accommodation due to a disability during the application or the recruiting process, please send a request via the Accommodation request form. Benefits/perks listed below may vary depending on the nature of your employment with Microsoft and the country where you work.
工作职责
Model Optimization & Deployment: Design and implement efficient workflows for training, distillation, and fine-tuning Small and Large Language Models (SLMs), leveraging techniques such as LoRA, QLoRA, and instruction tuning. Apply model compression strategies—including quantization (e.g., GPTQ, AWQ) and pruning—to reduce inference costs and improve latency. Optimize LLM inference performance using frameworks like vLLM and TensorRT-LLM (TRT-LLM) to enable scalable, low-latency deployment. Build robust and scalable inference systems tailored to heterogeneous production environments, with a strong focus on performance, cost-efficiency, and stability. Evaluation & Data Management： Develop evaluation datasets and metrics to assess model performance in real-world product scenarios. Build and maintain end-to-end machine learning pipelines encompassing data preprocessing, training, validation, and deployment. Cross-functional Collaboration： Collaborate closely with product managers, engineers, and research scientists to translate business needs into impactful AI solutions, driving real-world adoption and seamless product integration.
Bringing the State of the Art to Products Collaborates with and bridges the gap between researchers (in community, Microsoft Research [MSR], or in their own organizations) and development teams. Brings new technology and approaches into production by applying long-term research efforts to solve immediate product needs. With limited guidance from others, works to create product impact. Identifies approach, and applies, improves, or creates a research-backed solution (e.g., novel, data driven, scalable, extendable) to positively impact a Microsoft product or service. Solves components or aspects of a problem as assigned by a senior team member. May publish research to promote receiving new intellectual property for product impact. Participates in collaborative relationships with relevant product and business groups inside or outside of Microsoft and provides expertise or technology to create business impact. Participates in technology transfer attempts, filing patents, authoring white papers, developing or maintaining tools/services for internal Microsoft use, or consulting for product or business groups. May publish research to promote receiving new intellectual property for business impact. Capability Management and Networking Maintains ties with external network of peers and identifies prospective talent, when asked. May contribute to publications on research findings. May participate in candidate interviews. Collaborates with the academic community to develop the recruiting pipeline and establish awareness of their work. Reinforces a positive environment by applying best practices. May support mentorship by assisting with onboarding of research interns or other entry-level team members, if applicable. Documentation Performs documentation of work in progress, experimentation results, plans, etc. Documents scientific work to ensure process is captured. Participates in the creation of informal documentation and may share findings to promote innovation within group. Ethics and Privacy Understands and follows ethics and privacy policies when executing research processes and/or collecting data/information. Leveraging Applied Research Applies strategy by understanding the role in the team and applying the strategy provided by senior team members and incorporates state-of-the-art research. Asks probing questions to better understand strategy. Researches and develops an understanding of tools, technologies, and methods being used in the community that can be utilized to improve product quality, performance, or efficiency. Contributes knowledge around several specialized tools/methods to support the application of business impact or serves as an expert in a deeply specialized area. Gains deep knowledge in a service, platform, or domain and acquires knowledge of changes in industry trends and advances in applied technologies. Consults with engineers and product teams to apply advanced concepts to product needs. Learns product domain by reviewing products. Machine Learning Functionality, Insights, and Technical Tools Prepares data to be used for analysis by reviewing criteria that reflect quality and technical constraints. Reviews data and suggests data to be included and excluded. Describes actions taken to address data quality problems. Assists with the development of useable datasets for modeling purposes. Supports the scaling of feature ideation and data preparation. Helps take cleaned data and adapts for machine learning purposes, under the direction of a senior team member. Seeks guidance from senior team members when confronted with problems/challenges. Uses machine learning algorithms that structures, analyzes, and uses data in product and platforms to train algorithms for scalable artificial intelligence solutions before deploying. Begins to develop new machine learning improvements independently while under the direction of a senior team member. Collaborates to leverage data to identify pockets of opportunity to apply state-of-the-art algorithms to improve a solution to a business problem. Uses statistical analysis tools for evaluating Machine Learning models and validating assumptions about the data while also reviewing consistency against other sources. Begins to independently run basic descriptive, diagnostic, predictive, and prescriptive statistics. Assists with the communication of insights under the direction of senior team members. Supports the application and use of intelligence created during the training of algorithms for deployment. Seeks information about large-scale computing frameworks, data analysis systems, and modeling environments to improve models. Helps create a model, apply the model to real products, and then verify effects through iterations. Helps with experiments by putting multiple models in production and evaluating their performance. Sets up monitoring and implementation to track production models, under the direction of a senior team member. Addresses models when that break, under the direction of others. Leverages or designs and uses machine learning/data extraction, transformation, and loading (ETL) of pipelines (e.g., data collection, cleaning) based on data prepared.
A good candidate will play a key role in driving algorithmic and modeling improvement to the system, analyze performance and identify opportunities based on offline and online testing, develop and deliver robust and scalable solutions, make direct impact to both user and advertisers experience, and continually increase the revenue for Bing ads. The candidate should also have good communication, collaboration, and analytical skills.
职位：Applied scientist 应用科学家实习生 毕业时间：2025年10月 - 2026年9月之间毕业的应届毕业生 · 入职日期：2025年6月及之前 · 实习时间：保证一周实习4-5天全职实习，至少持续5个月 · 工作地点：北京朝阳区酒仙桥路恒通商务园区 · 校招信息请参考校园招聘申请手册： https://amazonexteu.qualtrics.com/CP/File.php?F=F_55YI0e7rNdeoB6e 投递须知： 1 填写简历申请时，请把必填和非必填项都填写完整。提交简历之后就无法修改了哦！ 2 学校的英文全称请准确填写。 如果您正在攻读计算机视觉、生成式AI或多模态领域的博士或硕士研究生，而且对应用科学家的实习工作感兴趣。 如果您也喜爱深入研究棘手的技术问题并提出解决方案，用成功的产品显著地改善人们的生活。 那么，我们诚挚邀请您加入亚马逊的International Technology自动化营销团队改善亚马逊节假日促销的用户体验。我们的目标是帮助亚马逊的客户找到他们所需的产品，并发现他们感兴趣的新产品。 这会是一份收获满满的工作。您每天的工作都与全球数百万亚马逊客户的体验紧密相关。您将提出和探索LLM和CV领域的创新，例如如何精准控制最前沿的基座大语言模型和图像生成模型以满足自动化的需求。您将集成这些模型到工具链中生成个性化的促销广告图，通过标注数据、建模和客户反馈来完成闭环。您对模型的选择需要能够平衡业务指标和响应时间的需求。
职位：Applied scientist 应用科学家实习生 毕业时间：2025年10月 - 2026年9月之间毕业的应届毕业生 · 入职日期：2025年6月及之前 · 实习时间：保证一周实习4-5天全职实习，至少持续3个月 · 工作地点：北京朝阳区酒仙桥路恒通商务园区 · 校招信息请参考校园招聘申请手册： https://amazonexteu.qualtrics.com/CP/File.php?F=F_55YI0e7rNdeoB6e 投递须知： 1 填写简历申请时，请把必填和非必填项都填写完整。提交简历之后就无法修改了哦！ 2 学校的英文全称请准确填写。 如果您正在攻读NLP，IR或搜索领域专业的博士或硕士研究生，而且对应用科学家的实习工作感兴趣。如果您也喜爱深入研究棘手的技术问题并提出解决方案，用成功的产品显著地改善人们的生活。 那么，我们诚挚邀请您加入亚马逊的International Technology搜索团队改善Amazon的产品搜索服务。我们的目标是帮助亚马逊的客户找到他们所需的产品，并发现他们感兴趣的新产品。 这会是一份收获满满的工作。您每天的工作都与全球数百万亚马逊客户的体验紧密相关。您将提出和探索NLP和IR领域的创新，基于TB级别的产品和流量数据设计机器学习模型。您将集成这些模型到搜索引擎中为客户提供服务，通过数据，建模和客户反馈来完成闭环。您对模型的选择需要能够平衡业务指标和响应时间的需求。