[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"me":3,"catalog:tr:data-engineer\u002Fpe-spark-api-semantics-udfs":4,"config":190},null,{"field_key":5,"field_name":6,"seniority":7,"topic_key":8,"topic_name":9,"spec_key":7,"spec_name":7,"locale":10,"cell_total":11,"field_total":12,"seniorities":13,"topics":17,"specs":94,"samples":105},"data-engineer","Data Engineer","","pe-spark-api-semantics-udfs","Pe Spark Api Semantics Udfs","tr",75,1950,[14,15,16],"junior","mid","senior",[18,21,24,27,30,33,36,39,42,45,48,51,52,55,58,61,64,67,70,73,76,79,82,85,88,91],{"key":19,"name":20,"count":11},"data-governance-lineage","Data Governance Lineage",{"key":22,"name":23,"count":11},"data-modeling-warehousing","Data Modeling Warehousing",{"key":25,"name":26,"count":11},"data-partitioning-scaling","Data Partitioning Scaling",{"key":28,"name":29,"count":11},"data-pipeline-design","Data Pipeline Design",{"key":31,"name":32,"count":11},"data-pipeline-orchestration","Data Pipeline Orchestration",{"key":34,"name":35,"count":11},"data-pipeline-reliability","Data Pipeline Reliability",{"key":37,"name":38,"count":11},"data-quality-validation","Data Quality Validation",{"key":40,"name":41,"count":11},"pe-airflow-scheduling-dependencies","Pe Airflow Scheduling Dependencies",{"key":43,"name":44,"count":11},"pe-dbt-models-materializations-tests","Pe Dbt Models Materializations Tests",{"key":46,"name":47,"count":11},"pe-file-formats-table-formats","Pe File Formats Table Formats",{"key":49,"name":50,"count":11},"pe-incremental-cdc-merge-mechanics","Pe Incremental Cdc Merge Mechanics",{"key":8,"name":9,"count":11},{"key":53,"name":54,"count":11},"pe-spark-execution-shuffle-partitions","Pe Spark Execution Shuffle Partitions",{"key":56,"name":57,"count":11},"st-flink-runtime-checkpointing-state","St Flink Runtime Checkpointing State",{"key":59,"name":60,"count":11},"st-flink-time-watermarks-windows","St Flink Time Watermarks Windows",{"key":62,"name":63,"count":11},"st-kafka-broker-storage-replication","St Kafka Broker Storage Replication",{"key":65,"name":66,"count":11},"st-kafka-connect-schema-registry","St Kafka Connect Schema Registry",{"key":68,"name":69,"count":11},"st-kafka-producer-consumer-semantics","St Kafka Producer Consumer Semantics",{"key":71,"name":72,"count":11},"streaming-fundamentals","Streaming Fundamentals",{"key":74,"name":75,"count":11},"st-streaming-joins-tables","St Streaming Joins Tables",{"key":77,"name":78,"count":11},"wh-bigquery-execution-slots-cost","Wh Bigquery Execution Slots Cost",{"key":80,"name":81,"count":11},"wh-bigquery-storage-partitioning-clustering","Wh Bigquery Storage Partitioning Clustering",{"key":83,"name":84,"count":11},"wh-materialized-views-caching","Wh Materialized Views Caching",{"key":86,"name":87,"count":11},"wh-redshift-distribution-sort-vacuum","Wh Redshift Distribution Sort Vacuum",{"key":89,"name":90,"count":11},"wh-snowflake-compute-warehouses","Wh Snowflake Compute Warehouses",{"key":92,"name":93,"count":11},"wh-snowflake-storage-micropartitions","Wh Snowflake Storage Micropartitions",[95,99,102],{"key":96,"name":97,"count":98},"pipelines-etl","Pipelines \u002F ETL",450,{"key":100,"name":101,"count":98},"streaming","Streaming",{"key":103,"name":104,"count":98},"warehousing","Warehousing",[106,124,138,151,164,177],{"id":107,"topic":9,"difficulty":108,"body":109,"options":110,"correct_key":115,"explanation":123},"01a00ae7-b517-744b-a4ab-e069ed0e13ac",2,"Spark'ın Catalyst optimize edicisinde bir DataFrame sorgusu çalıştırılmadan önce birkaç plan aşamasından geçer. Doğru sıralama hangisidir?",[111,114,117,120],{"key":112,"text":113},"a","Fiziksel plan → optimize edilmiş mantıksal plan → mantıksal plan",{"key":115,"text":116},"b","Mantıksal plan → optimize edilmiş mantıksal plan → fiziksel plan",{"key":118,"text":119},"c","Optimize edilmiş mantıksal plan → mantıksal plan → fiziksel plan",{"key":121,"text":122},"d","Fiziksel plan → mantıksal plan → optimize edilmiş mantıksal plan","Catalyst önce DataFrame\u002FSQL kodundan mantıksal planı kurar, kural tabanlı optimizasyonlar uygulayarak optimize edilmiş mantıksal planı üretir, son olarak çalıştırma motorunun kullandığı fiziksel planı üretir.",{"id":125,"topic":9,"difficulty":126,"body":127,"options":128,"correct_key":112,"explanation":137},"01a00ae7-b518-7626-abe4-2912eea0e00d",1,"```python\ndf.filter(df.amount > 100).explain()\n```\nBir DataFrame üzerinde argümansız `.explain()` çağrısı varsayılan olarak hangi planı basar?",[129,131,133,135],{"key":112,"text":130},"Yalnızca fiziksel planı",{"key":115,"text":132},"Yalnızca çözülmemiş mantıksal planı",{"key":118,"text":134},"Dört plan aşamasının tamamını (parsed, analyzed, optimized, physical)",{"key":121,"text":136},"Yalnızca optimize edilmiş mantıksal planı","Argümansız (extended=False) `.explain()` yalnızca fiziksel planı basar. extended=True ya da 'formatted' gibi bir mod string'i verilirse mantıksal plan aşamaları dahil daha fazla detay gösterilir.",{"id":139,"topic":9,"difficulty":108,"body":140,"options":141,"correct_key":121,"explanation":150},"01a00ae7-b519-7ba9-bf03-b488d145c498","Catalyst'in optimize edilmiş mantıksal planında 'column pruning' (sütun budama) ne yapar?",[142,144,146,148],{"key":112,"text":143},"Bir filtre koşulunu geçemeyen satırları scan öncesi kaldırır",{"key":115,"text":145},"Aynı tablonun iki scan'ini tek scan'de birleştirir",{"key":118,"text":147},"Join işlemlerini tahmini maliyete göre yeniden sıralar",{"key":121,"text":149},"Downstream'de referans edilmeyen sütunları düşürür","Column pruning, plan içinde daha sonra gerçekten kullanılan sütunları bulur ve scan'i yalnızca o sütunları okuyacak şekilde yeniden yazar; özellikle kolon tabanlı formatlarda I\u002FO'yu azaltır.",{"id":152,"topic":9,"difficulty":108,"body":153,"options":154,"correct_key":118,"explanation":163},"01a00ae7-b51c-7676-a838-ea98052a861d","Catalyst'in constant folding optimizasyonu `WHERE price > 10 + 5` gibi bir ifadeyle ne yapar?",[155,157,159,161],{"key":112,"text":156},"Tüm ifadeyi olduğu gibi depolama katmanına gönderir",{"key":115,"text":158},"Karşılaştırmayı bir join koşuluna çevirir",{"key":118,"text":160},"`10 + 5`'i plan zamanında hesaplayıp `15` ile değiştirir",{"key":121,"text":162},"Filtreyi bir window fonksiyonuna dönüştürür","Constant folding, yalnızca literallerden oluşan ifadeleri sorgu planlama sırasında hesaplar; böylece fiziksel plan her satır için `10 + 5`'i yeniden hesaplamak yerine doğrudan `price > 15` karşılaştırması yapar.",{"id":165,"topic":9,"difficulty":108,"body":166,"options":167,"correct_key":115,"explanation":176},"01a00ae7-b51f-7546-9843-17734e1bc90a","Spark filtrelenebilir bir veri kaynağından okurken 'predicate pushdown' ne anlama gelir?",[168,170,172,174],{"key":112,"text":169},"Filtreler join aşamasından önce değil sonra çalışacak şekilde taşınır",{"key":115,"text":171},"Filtre kaynağa iletilir, kaynak eşleşmeyen veriyi erkenden eleyebilir",{"key":118,"text":173},"Filtreler esneklik için UDF'e çevrilir",{"key":121,"text":175},"Tüm filtreler yalnızca driver'da değerlendirilir","Predicate pushdown ile Spark, filtre koşulunu kaynağın (JDBC, Parquet vb.) kendisinin uygulayabileceği bir forma çevirir; böylece cluster'a fiilen okunan satır\u002Fdosya sayısı azalır.",{"id":178,"topic":9,"difficulty":108,"body":179,"options":180,"correct_key":121,"explanation":189},"01a00ae7-b521-77f8-91fb-99e733a4d18a","Bir ekip şunu yazıyor:\n```python\ndf.filter(is_valid_udf(df.email))\n```\n`is_valid_udf` bir Python UDF. `explain()` çıktısına baktıklarında altta yatan Parquet kaynağına pushdown konusunda ne beklemeliler?",[181,183,185,187],{"key":112,"text":182},"Parquet keyfi filtreleri desteklediği için UDF filtresi otomatik pushdown edilir",{"key":115,"text":184},"Catalyst pushdown'dan önce UDF'i built-in bir ifadeye çevirir",{"key":118,"text":186},"Catalyst UDF bytecode'unu inline ettiği için pushdown yine gerçekleşir",{"key":121,"text":188},"UDF filtresi pushdown edilemez; Catalyst onu scan'den sonra değerlendirir","Catalyst yalnızca ifade olarak anlayabildiği filtreleri pushdown edebilir. UDF optimize edici için opaktır; içinde bulunduğu filtre satırlar okunduktan sonra Spark içinde değerlendirilir, Parquet okuyucusuna itilmez.",{"fields":191,"seniorities":408,"interview_shapes":409,"locales":414,"oauth":416,"question_count":419,"coach_enabled":420,"jd_match_enabled":420},[192,217,237,254,278,291,310,329,351,370,377,399],{"key":193,"name_tr":194,"name_en":194,"sort":126,"specializations":195},"backend","Backend",[196,199,202,205,208,211,214],{"key":197,"name":198,"field":193},"general","Genel",{"key":200,"name":201,"field":193},"go","Go",{"key":203,"name":204,"field":193},"python","Python",{"key":206,"name":207,"field":193},"java","Java",{"key":209,"name":210,"field":193},"csharp","C#\u002F.NET",{"key":212,"name":213,"field":193},"nodejs","Node.js",{"key":215,"name":216,"field":193},"php","PHP",{"key":218,"name_tr":219,"name_en":219,"sort":108,"specializations":220},"frontend","Frontend",[221,222,225,228,231,234],{"key":197,"name":198,"field":218},{"key":223,"name":224,"field":218},"javascript","JavaScript",{"key":226,"name":227,"field":218},"typescript","TypeScript",{"key":229,"name":230,"field":218},"react","React",{"key":232,"name":233,"field":218},"vue","Vue",{"key":235,"name":236,"field":218},"angular","Angular",{"key":238,"name_tr":239,"name_en":239,"sort":240,"specializations":241},"fullstack","Fullstack",3,[242,243,244,245,246,247,248,249,250,251,252,253],{"key":197,"name":198,"field":238},{"key":200,"name":201,"field":193},{"key":203,"name":204,"field":193},{"key":206,"name":207,"field":193},{"key":209,"name":210,"field":193},{"key":212,"name":213,"field":193},{"key":215,"name":216,"field":193},{"key":223,"name":224,"field":218},{"key":226,"name":227,"field":218},{"key":229,"name":230,"field":218},{"key":232,"name":233,"field":218},{"key":235,"name":236,"field":218},{"key":255,"name_tr":256,"name_en":256,"sort":257,"specializations":258},"devops-cloud","DevOps \u002F Cloud",4,[259,260,263,266,269,272,275],{"key":197,"name":198,"field":255},{"key":261,"name":262,"field":255},"aws","AWS",{"key":264,"name":265,"field":255},"gcp","GCP",{"key":267,"name":268,"field":255},"azure","Azure",{"key":270,"name":271,"field":255},"kubernetes","Kubernetes",{"key":273,"name":274,"field":255},"terraform","Terraform",{"key":276,"name":277,"field":255},"linux","Linux",{"key":279,"name_tr":280,"name_en":280,"sort":281,"specializations":282},"ai-engineer","AI Engineer",5,[283,284,285,288],{"key":197,"name":198,"field":279},{"key":203,"name":204,"field":279},{"key":286,"name":287,"field":279},"llm-rag","LLM\u002FRAG",{"key":289,"name":290,"field":279},"mlops","MLOps",{"key":292,"name_tr":293,"name_en":294,"sort":295,"specializations":296},"database","Veritabanı","Database",6,[297,298,301,304,307],{"key":197,"name":198,"field":292},{"key":299,"name":300,"field":292},"postgresql","PostgreSQL",{"key":302,"name":303,"field":292},"mysql","MySQL",{"key":305,"name":306,"field":292},"mongodb","MongoDB",{"key":308,"name":309,"field":292},"redis","Redis",{"key":311,"name_tr":312,"name_en":313,"sort":314,"specializations":315},"mobile","Mobil","Mobile",7,[316,317,320,323,326],{"key":197,"name":198,"field":311},{"key":318,"name":319,"field":311},"ios-swift","iOS (Swift)",{"key":321,"name":322,"field":311},"android-kotlin","Android (Kotlin)",{"key":324,"name":325,"field":311},"flutter","Flutter",{"key":327,"name":328,"field":311},"react-native","React Native",{"key":330,"name_tr":331,"name_en":332,"sort":333,"specializations":334},"security","Güvenlik","Security",8,[335,336,339,342,345,348],{"key":197,"name":198,"field":330},{"key":337,"name":338,"field":330},"appsec","AppSec",{"key":340,"name":341,"field":330},"offensive-pentest","Offensive \u002F Pentest",{"key":343,"name":344,"field":330},"cloud-security","Cloud Security",{"key":346,"name":347,"field":330},"devsecops","DevSecOps",{"key":349,"name":350,"field":330},"blue-team-incident","Blue Team \u002F Incident",{"key":352,"name_tr":353,"name_en":354,"sort":355,"specializations":356},"qa-test-automation","QA \u002F Test Otomasyonu","QA \u002F Test Automation",9,[357,358,361,364,367],{"key":197,"name":198,"field":352},{"key":359,"name":360,"field":352},"test-automation","Test Automation",{"key":362,"name":363,"field":352},"sdet","SDET",{"key":365,"name":366,"field":352},"performance-testing","Performance Testing",{"key":368,"name":369,"field":352},"mobile-qa","Mobile QA",{"key":5,"name_tr":6,"name_en":6,"sort":371,"specializations":372},10,[373,374,375,376],{"key":197,"name":198,"field":5},{"key":96,"name":97,"field":5},{"key":100,"name":101,"field":5},{"key":103,"name":104,"field":5},{"key":378,"name_tr":379,"name_en":380,"sort":381,"specializations":382},"game-dev","Oyun Geliştirme","Game Development",11,[383,384,387,390,393,396],{"key":197,"name":198,"field":378},{"key":385,"name":386,"field":378},"unity-csharp","Unity (C#)",{"key":388,"name":389,"field":378},"unreal-cpp","Unreal (C++)",{"key":391,"name":392,"field":378},"gameplay","Gameplay",{"key":394,"name":395,"field":378},"graphics-rendering","Graphics \u002F Rendering",{"key":397,"name":398,"field":378},"multiplayer-netcode","Multiplayer \u002F Netcode",{"key":400,"name_tr":401,"name_en":401,"sort":402,"specializations":403},"ml-engineer","ML Engineer",12,[404,405],{"key":197,"name":198,"field":400},{"key":406,"name":407,"field":400},"classical-ml","Classical ML",[14,15,16],{"junior":410,"mid":412,"senior":413},{"questions":411,"median_sec":3},20,{"questions":411,"median_sec":3},{"questions":411,"median_sec":3},[10,415],"en",[417,418],"google","github",27600,true]