Infra Reliability Engineer: AI Data Pipelines, Remote

Bright Vision Technologies; United States; Remote; USD 125,000 - 170,000; Full time

Job summary

Bright Vision Technologies is seeking an Infrastructure Reliability Engineer to build and operate the large-scale data systems that power AI training and evaluation pipelines.

The role combines data engineering expertise with a strong understanding of AI workloads, focusing on ingestion, transformation, quality assurance, lineage, and high-throughput delivery of data to training jobs across diverse modalities.

Qualifications

• Bachelor’s or Master’s degree in Computer Science or a related field.
• Six or more years of data engineering experience, with significant work supporting ML or AI workloads.
• Strong proficiency in Python and at least one JVM or systems language.
• Deep experience with modern data processing frameworks such as Spark, Ray, or Beam.
• Hands-on experience operating petabyte-scale storage and pipeline systems.
• Strong understanding of distributed systems, data modeling, and storage formats.
• Experience with dataset versioning, lineage, and reproducibility for ML workflows.
• Familiarity with high-throughput data loading for accelerator-based training.
• Strong software engineering practices including testing, CI/CD, and code review.
• Excellent communication and cross-functional collaboration skills.

Responsibilities

• Design and operate large-scale data pipelines supporting AI training, evaluation, and continual improvement workflows.
• Build ingestion systems for diverse modalities including text, image, audio, video, and structured signals.
• Implement data cleaning, deduplication, filtering, and quality assurance at petabyte scale.
• Develop dataset versioning, lineage, and provenance tracking systems suitable for reproducible training.
• Build high-throughput data loading systems that maximize GPU utilization during training.
• Implement labeling workflows, active learning pipelines, and human-in-the-loop data improvement systems.
• Design storage architectures balancing cost, throughput, and latency across data tiers.
• Build evaluation dataset construction pipelines with strict integrity and contamination controls.
• Implement data privacy, redaction, and consent enforcement throughout the pipeline.
• Collaborate with ML researchers and engineers to align data systems with model development needs.
• Drive observability of data quality, drift, and pipeline health across the AI data estate.
• Optimize cost and performance through compression, format selection, and caching strategies.
• Document data systems, schemas, and operational procedures for broad internal use.
• Stay current with AI data infrastructure research and emerging open-source tools.

Skills

Python

JVM language

Distributed systems

CI/CD

Education

Bachelor’s or Master’s degree in Computer Science

Tools

Spark

Ray

Beam

Job description

Bright Vision Technologies is seeking an Infrastructure Reliability Engineer to build and operate the large-scale data systems that power AI training and evaluation pipelines.

The role combines data engineering expertise with a strong understanding of AI workloads, focusing on ingestion, transformation, quality assurance, lineage, and high-throughput delivery of data to training jobs across diverse modalities.