Data Engineer, Machine Learning

Sesame • San Francisco, CA

1d • $170,000 - $240,000

About The Position

Sesame is seeking a Data Engineer to construct and manage the data pipelines essential for feeding the company's AI models. This role involves close collaboration with machine learning engineers and researchers, ensuring they have access to data in the correct format and at the opportune moments for model training, evaluation, and deployment. The data at Sesame is diverse and intricate, encompassing conversations, voice recordings, sensor signals, and product telemetry. The engineer will be responsible for designing systems that transform raw, unstructured, multimodal data into clean, version-controlled, and well-documented datasets that ML teams can rely on. This is a highly technical, infrastructure-centric position, more aligned with ML engineering than traditional data analytics. The role requires deep integration with ML teams to understand their workflows and to build infrastructure that accelerates the entire model development lifecycle, from data collection and labeling to training and evaluation.

Requirements

• 5+ years in data engineering, with meaningful experience supporting ML or AI teams specifically.
• Strong SQL and Python skills — you'll use both daily.
• Experience building and operating ETL/ELT pipelines at scale using modern data platforms and tooling.
• Experience with workflow orchestration systems such as Airflow, Dagster, or Prefect.
• Hands-on experience with ML data workflows: training data pipelines, dataset versioning, data labeling pipelines, or model evaluation data.
• A solid understanding of how ML teams work — you don't need to train models; what matters is understanding what makes a good training dataset and why data quality directly affects model performance.
• Comfort working with unstructured and semi-structured data — audio, text, JSON logs — not just clean relational tables.
• Strong communication skills. You'll be embedded with ML engineers and need to bridge data systems and model requirements effectively.

Nice To Haves

• Vector databases, embedding storage, or feature stores.
• Data from hardware or embedded systems: telemetry, sensors, real-time streams.
• Distributed compute frameworks for large-scale data processing such as Ray or Spark.
• Kubernetes and managed Kubernetes environments such as GKE or EKS.
• Data privacy frameworks, especially around voice or conversational data.
• Building internal tooling or self-serve data platforms.

Responsibilities

• Design and build production data pipelines that prepare conversational, voice, and multimodal data for model training and evaluation.
• Partner directly with ML engineers to understand data requirements for new models and experiments, and deliver datasets that meet those needs.
• Build and maintain infrastructure for dataset versioning, lineage tracking, and reproducibility — so any training run can be traced back to its exact data.
• Develop data quality frameworks that catch issues before they become model quality issues: schema validation, drift detection, and coverage monitoring.
• Optimise large-scale data processing for cost and performance across Sesame's cloud infrastructure.
• Build tooling that makes it easy for ML engineers and researchers to discover, explore, and request data independently.
• Define and enforce data governance and privacy standards, particularly around sensitive conversational and voice data.
• Contribute to architecture decisions around Sesame's broader data platform as the team and data volume grow.

Benefits

• 401 (k) max employer match: 3.5% of compensation
• 100% employer-paid health, vision, and dental benefits for you and your dependents
• Unlimited PTO and sick time
• Flexible spending account with employer matching up to $1,650/year (medical FSA)
• Guardian Employee Assistance Program (EAP)
• Opportunity to share in the company's success with competitive stock options

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed