Learning Strategies and Interview Preparation Insights from a Big Data Student

Late‑night coding sessions are recorded here, focusing on a junior student (nicknamed "Lucky Sister") from a big‑data intensive class, whose background is modest but whose learning approach offers valuable lessons.

Attitude toward learning: Despite widespread complacency, Lucky Sister consistently stayed up late during project sprints, often outworking the instructor, demonstrating that personal effort drives better opportunities.

Learning summaries: Detailed notes are required for every project; the instructor has accumulated over 5 million characters of notes, emphasizing that thorough documentation becomes a crucial interview resource. The class’s framework and project notes, contributed by mentors and peers, help students master difficult topics and excel in technical interviews across Chinese companies.

Ask questions actively: The instructor shares screenshots of classmates’ questions, encouraging learners to voice doubts promptly.

Emphasize expression and communication: Structured presentation of project experience is vital; the article includes an example of a well‑crafted project description image.

Finally, a collection of interview questions (B‑side) is provided for reference:

第一轮<br/>1.介绍项目，项目中的重点难点<br/>2.hive的优化，这个好几家公司都问了<br/>3.hive sql的执计划<br/>4.hive和mysql的区别<br/>5.Sort by 和order by的区别<br/>6.数据倾斜的场景，如何解决的<br/>7.sql题<br/>字段:订单id，时间，用户id<br/>计算10分钟内连续下单大于100次的用户<br/>第二轮<br/>1.介绍项目，项目中的重点难点<br/>2.数仓建模理论<br/>3.冷热数据如何处理<br/>4.数据治理从哪几个方面进行<br/>5.数据质量的衡量标准，数据质量的效果，如何验收，项目流程<br/>6.用的星型还是雪花模型，区别是什么？<br/>第三轮<br/>1.介绍项目，项目中的重点难点<br/>2.linux命令 查找文件，awk命令<br/>3.kafka分区，ack机制<br/>4.spark的执行原理<br/>5.解析下spark的DAG<br/>6.mr的执行原理<br/>7.大小表join的优化<br/>8.Spark常用算子reduceByKey与groupByKey的区别，哪一种更具优势?<br/>9.Spark任务执行模式，提交任务，资源也够的情况下，还是不能跑，啥原因 <br/>10.spark和MR的区别<br/>第四轮<br/>1.介绍项目，项目中的重点难点<br/>2.项目中遇到啥问题<br/>3.kafka丢失数据，怎么解决<br/>4.kafka的核心组件介绍 topic，broker，partition，consumer，producer<br/>5.clickhouse的各类引擎，怎么用的，啥原理，你们咋用的<br/>6.Flink checkpoint执行流程<br/>7.flink和spark对比<br/>第五轮<br/>1.介绍项目，项目中的重点难点<br/>2.数据中台oneid，oneservice<br/>3.遇到啥问题，项目进度把控，资源协调<br/>4.数据的安全，权限的管理<br/>5.数仓重构，数仓模型的建设,遇到啥问题，什么样的周期，如何安排的，效率咋样?<br/>

For further resources, the article links to a massive 3‑million‑character big‑data interview community and a list of related articles covering Hive, Spark, Flink, ClickHouse, data governance, and career growth.