Senior Data Engineer

Role Summary

We are looking for a seasoned Senior Data Engineer to architect, build, and own the data pipelines that power our large language model (LLM) development. As a senior Individual Contributor (IC), you will be the team's expert on data ingestion, processing, and quality for all AI training.

Your primary mission is to build scalable, automated systems that transform massive, raw datasets into pristine, model-ready formats. While your focus will be on data engineering, your expertise will be valued in collaborating on model training runs and experiments. You're the perfect fit if you are a Python expert who thrives on solving large-scale data challenges and enjoys working at the intersection of data engineering and machine learning.

Key Responsibilities

Architect & Build: Design, develop, and own robust, scalable, and automated ETL/ELT pipelines in Python for ingesting and processing terabyte-scale text datasets.

Data Quality: Implement rigorous data cleaning, deduplication, filtering, and normalization strategies. Define and enforce data quality standards to ensure the highest integrity for model training.

Data Transformation

Collaboration: Work closely with our team of AI researchers and ML engineers to understand data requirements, define metrics, and support the model training lifecycle.

Optimization: Continuously optimize data processing workflows for speed, cost, and reliability.

ML Support (Secondary):

Required Qualifications

8+ years of professional experience in data engineering, data processing, or backend software engineering.

Expert-level proficiency in Python and its data ecosystem (e.g., Pandas, NumPy, Dask, Polars).

Proven experience building and maintaining large-scale data pipelines.

Deep understanding of data structures, data modeling, and software engineering best practices (Git, CI/CD, testing).

Experience handling and parsing diverse data formats (JSON, CSV, XML, Parquet) at scale.

Excellent problem-solving skills and a meticulous attention to detail.

Strong communication and collaboration skills, with experience working in a team environment.

Preferred Qualifications (Nice-to-Haves)

Hands-on experience with the data preprocessing pipeline for an LLM (e.g., LLaMA, BERT, GPT-family).

Strong experience with big data frameworks like Apache Spark or Ray.

Experience with Hugging Face libraries (Transformers, Datasets, Tokenizers).

Familiarity with ML frameworks like PyTorch or TensorFlow.

Proficiency with cloud platforms (AWS, GCP, Azure) and their data/storage services.