AI Data Engineer: Scale Data Pipelines for Training

Speechify; United States; Remote; USD 140,000 - 200,000; Full time

Benefits offered by this job

Competitive compensation

Autonomy

Asynchronous work culture

Job summary

Speechify is seeking a Software Engineer - AI Data to join our AI team and build data pipelines powering model training and next-generation products. We create high-quality datasets at petabyte scale by combining engineering, infrastructure, and research to keep AI development fast and cost-effective.

You will discover new audio data sources, build scalable pipelines, and collaborate with AI scientists to improve training sets and support models.

Qualifications

• Degree in Computer Science or related field or equivalent practical experience.
• 5+ years of professional software engineering with backend focus.
• Strong Python and Bash proficiency in Linux environments.
• Experience with data pipelines or large-scale data processing.
• Ability to work independently in ambiguous environments.

Responsibilities

• Discover new sources of audio data and build pipelines to ingest them into training datasets.
• Build and optimize large-scale data-processing systems for quality, throughput and cost.
• Collaborate with AI Scientists to improve training datasets and model development.
• Own engineering projects from discovery through implementation and iteration.

Skills

Python

Bash

Linux environments

Data pipelines

Production systems

Independent worker

Education

Bachelor's in Computer Science or related field

Tools

Docker

Terraform

GCP

Job description

Speechify is seeking a Software Engineer - AI Data to join our AI team and build data pipelines powering model training and next-generation products. We create high-quality datasets at petabyte scale by combining engineering, infrastructure, and research to keep AI development fast and cost-effective.

You will discover new audio data sources, build scalable pipelines, and collaborate with AI scientists to improve training sets and support models.