Job Description & Details
This is a hands-on hybrid role bridging heavy-duty data engineering with bleeding-edge Generative AI development, operating out of either Reston, VA or Washington, DC. You will be building the underlying data pipelines that feed modern LLM applications while wrestling with real-world enterprise data hygiene and cloud architectures.
What You'll Actually Be Doing
Your days will be split between traditional data engineering and modern AI implementation. You will write clean Python and optimized SQL to build ETL pipelines, but instead of just dumping tables into a dashboard, you are prepping that data for vector search, embeddings, and RAG architectures. Expect to spend a lot of time integrating LLMs via APIs like Amazon Bedrock, tuning prompts, and figuring out why your model is hallucinating on edge-case enterprise data.
The Core Tech Stack
Python and advanced SQL are non-negotiable here; if you can't write performant queries or manage data transformations, you'll sink. You also need solid AWS experience, specifically touching serverless and data services like AWS Lambda and Amazon Redshift, alongside a firm grasp of vector databases, semantic search, and prompt engineering frameworks.
Interview Expectations
Be ready for deep-dive technical questions regarding how you handle pipeline failures during high-latency LLM API calls and how you optimize slow SQL queries against massive unstructured datasets. The interviewer is secretly testing your pragmatic problem-solving skills—they want to see how you balance the probabilistic nature of AI with the deterministic requirements of enterprise data pipelines.
Application Advice
Make sure your resume heavily features the exact overlap of Python, AWS, and LLM implementations like RAG or Amazon Bedrock. Avoid vague terms and spell out your experience with vector databases and prompt engineering so you easily clear their automated ATS filters.