Senior Platform Engineer, Khan Academy

• Bachelor’s or Master’s degree in Computer Science, Data Engineering, or a related field
• 5 years of Software Engineering experience with 3+ of those years working with large ML datasets, especially those in open-source repositories such as Hugging Face
• Strong programming skills in Go, Python, SQL, and at least one data pipeline framework (e.g., Airflow, Dagster, Prefect)
• Experience with data versioning tools (e.g., DVC, LakeFS) and cloud storage systems
• Familiarity with machine learning workflows — from training data preparation to evaluation
• Familiarity with the architecture and operation of large language models, and a nuanced understanding of their capabilities and limitations
• Attention to detail and an obsession with data quality and reproducibility
• Motivated by the Khan Academy mission “to provide a free world-class education for anyone, anywhere."
• Proven cross-cultural competency skills demonstrating self-awareness, awareness of other, and the ability to adopt inclusive perspectives, attitudes, and behaviors to drive inclusion and belonging throughout the organization

Desirable

• We’re looking for an ML Data Engineer to evolve our eval dataset tools to meet the growing platform needs of AI-based tutoring at Khan Academy
• We’re looking for someone who can gather internal requirements, design schema based on well-known dataset patterns, and deploy, document, and train people on an internal dataset management framework
• The systems you design will need to integrate with trace management and human labeling APIs
• You’ll work closely with other AI engineers, platform developers, and labeling teams to ensure our data is clean, representative, and ready for both human and automated evaluation
• This role bridges ML operations, data engineering and data science— enabling our AI systems to learn from reliable, well-structured datasets that reflect the diversity and nuance of real learners
• Evolve and maintain pipelines for transforming raw trace data into ML-ready datasets
• Clean, normalize, and enrich data while preserving semantic meaning and consistency
• Prepare and format datasets for human labeling, and integrate results into ML datasets
• Develop and maintain scalable ETL pipelines using Airflow, DBT, Go, and Python running on GCP
• Implement automated tests and validation to detect data drift or labeling inconsistencies
• Collaborate with AI engineers, platform developers, and product teams to define data strategies in support of continuously improving the quality of Khan’s AI-based tutoring
• Contribute to shared tools and documentation for dataset management and AI evaluation
• Inform our data governance strategies for proper data retention, PII controls/scrubbing, and isolation of particularly sensitive data such as offensive test imagery