I went through four interview rounds at Altimetrik for a Data Engineer position. The first was a technical round lasting about 45 minutes, where I was asked about window functions in Spark Scala and scenario-based questions around data engineering concepts. The interviewer also probed into Spark optimization techniques and how I'd applied them in past projects. The second round was managerial, similarly structured around 45–50 minutes, and focused on my background and approach to technical problems. Then came a client interview, which felt like the most intense round—they asked very specific, hands-on questions. One question asked me to write PySpark code for an SCD2 (Slowly Changing Dimension) scenario, which required translating business logic into actual transformation logic. They also drilled into SQL coding problems and Python fundamentals like writing a factorial series script. A lot of time was spent discussing the actual project I'd be working on and how Spark would fit into the ETL pipeline. The final round was with HR, which was more about logistics and fit. Throughout the process, I was tested on SQL, Python, and Scala/PySpark pretty heavily, with questions on query optimization, monitoring, ETL versus ELT differences, and JOIN patterns. One interviewer asked me to write a query using CTEs for a specific scenario. There were also deeper dives into shallow versus deep copy concepts in Python. The entire process moved quickly—I cleared all four rounds and was asked to upload documents and complete Digi verification, which felt like a strong signal. However, the offer was delayed significantly without clear communication from HR, and when it finally came, I had to decline due to personal circumstances. The experience was solid technically, but the lack of clarity around mandatory skill requirements upfront and the HR communication delays were frustrating. If you're interviewing here, come prepared with hands-on PySpark code examples and be ready to translate business problems into spark transformations, not just theoretical knowledge.
Prep tip from this candidate
Focus heavily on hands-on PySpark coding, especially scenario-based problems like SCD2 transformations and ETL logic. Be prepared to write actual code, not just explain concepts. Also confirm at the screening stage whether PySpark is a hard requirement for your target project—this will save time if it's not a strong area.
