Back to Jobs

AWS Data Platform Specialist

Not Disclosed

Job Description & Details

This role is a senior‑level AWS data platform specialist based in St. Louis, focusing on building end‑to‑end data pipelines on the AWS stack. If you enjoy turning raw logs into reliable analytics and have a telecom background, this could be a good fit.

What You'll Actually Be Doing

You’ll spend most of your day stitching together S3 buckets, Glue jobs, and Redshift clusters to move terabytes of data from source to analytics. Expect to design both batch and ELT pipelines, write performant SQL for data modeling, and use Lambda + CloudWatch for orchestration and monitoring. The job will also require you to troubleshoot schema drift and keep the pipelines cost‑effective.

The Core Tech Stack

The non‑negotiables are deep experience with AWS S3, Glue, Redshift, Lambda, and CloudWatch, plus solid SQL development and data‑modeling chops. The company needs you to own the whole data flow—from raw ingestion in S3, through transformation in Glue, to loading and serving in Redshift—so any gaps in those services will quickly become blockers.

Interview Expectations

  1. Design a near‑real‑time pipeline: Explain how you’d ingest JSON files from S3, transform them with Glue, and load into Redshift while handling schema evolution and ensuring low latency. The interviewer is looking for your ability to balance consistency, performance, and cost, plus how you’d use versioned schemas or Glue job bookmarks.
  2. Lambda‑CloudWatch error handling: Describe a situation where you used Lambda functions triggered by CloudWatch events to catch ETL failures, what metrics you’d surface (e.g., duration, error count, DLQ size), and how you’d set up alerts. They want to see you can build resilient, observable pipelines, not just fire‑and‑forget jobs.

Application Advice

Tailor your resume to hit the exact buzzwords the JD repeats: AWS S3, AWS Glue, AWS Redshift, ETL/ELT pipelines, data modeling, SQL development, AWS Lambda, CloudWatch, and if you have any telecom domain projects, surface them front‑and‑center. Highlight 8‑10 years of end‑to‑end data platform work and quantify impact (e.g., “Reduced pipeline latency by 30%”). Use the phrase “Face‑to‑Face interview” only if you’re comfortable with onsite work, as the posting explicitly mentions it.