over 1 year ago
Base Salary
$170k - $230k/yr
Responsibilities
- Architect and build workflows for ingesting, processing, and preparing multimodal training data for distributed ML training on GPU clusters.
- Develop compliant ingestion strategies for internal databases, third-party APIs, public biomedical datasets, and Tempus proprietary data.
- Use, optimize, and contribute to large-scale ML data loading and streaming frameworks.
- Collaborate with infrastructure and platform teams to optimize GCP cloud-native services for performance, cost, and security.
- Build connectors and data loaders for knowledge graphs, structured databases, biomedical literature repositories such as PubMed, and curated ontologies.
- Optimize storage for large-scale model training and knowledge access.
- Orchestrate, monitor, and troubleshoot data workflows with Airflow and Kubeflow Pipelines.
- Establish monitoring, logging, alerting, data drift detection, and data quality systems for pipeline health.
- Analyze and improve data I/O performance across storage, network, and compute resources.
- Manage and optimize cloud costs associated with storing and processing massive datasets.
Requirements
- Master’s degree in Computer Science, Artificial Intelligence, Software Engineering, or a related field.
- 8+ years of industry experience designing, building, and operating large-scale production data pipelines and infrastructure.
- Strong experience with massive heterogeneous datasets of terabytes or more and distributed data processing tools such as Apache Spark, Ray, or Dask.
- Hands-on experience with large-scale ML data frameworks such as Hugging Face Datasets, MosaicML Streaming, WebDataset, or Petastorm.
- Experience with MLOps tools and platforms such as MLflow, Kubeflow, or SageMaker Pipelines.
- Understanding of data challenges involved in training foundation models, LLMs, and multimodal models.
- Proficiency in Python and modern distributed data processing frameworks.
- Ability to influence technical direction and data strategy, mentor junior engineers, and collaborate cross-functionally.
- Excellent communication skills and ability to explain complex technical concepts to diverse audiences.
- Preferred: PhD or other advanced degree in Computer Science, Engineering, Bioinformatics, or a related field.
- Preferred: Contributions to relevant open-source projects.
- Preferred: Direct experience with clinical or biological data, including EHR, genomics, or medical imaging.
Benefits
- Tempus offers a full range of benefits, which may include medical and other benefits depending on the position.
- Benefits may include restricted stock units and incentive compensation.
- The role may be performed remotely in the USA or from New York, California, or Illinois, with location-specific salary ranges.
Tech Stack
Categories
Data EngineeringML Engineering
About Tempus
Tempus builds AI-enabled precision medicine tools and genomic tests used by physicians and health systems, with a core focus on oncology. It sells diagnostics, clinical decision-support software, and real‑world data services to providers and biopharma for drug discovery and trials. Founded in 2015 and headquartered in Chicago, Tempus is a public company listed on Nasdaq.