DataOps Engineer, Synthesia

• 3+ years minimum experience in Data Engineering / Data Ops / Data Science
• Been involved in managing large scale datasets not just one-off data collection tasks, you have seen continuous data collection
• Been responsible for setting up data ops (ingest / storage / transform / access) end-to-end for multiple teams
• Seen audio/video data and understand managing audio/video data at PB scale
• Experience with Streaming / Batch Data Pipelines (Airflow, Apache Beam, Spark etc.)
• Experience with event-driven systems
• Experience in handling heterogeneous types of data (e.g. audio / text / video / tabular data)
• Experience with any type of RDBMs (Relational Database Management Systems)
• Outstanding communication skills

Desirable

• Data Ops for data management, versioning, usage tracking, logging
• Setup of a data-lake and data transform pipelines for large scale audio-visual datasets
• Integration of 3d party annotation services for continuous data annotation and active learning
• Setup of metadata stores and APIs to access data-sets on demand for ML training
• Support for data streaming to train large models
• Data pipelines - deploy custom ML data transformations, working with our ML team
• Data access - create transient data-sets on demand to support ML model training
• Data tracking - usage tracking and monitoring across all data sources
• Establish the workflow for continual data delivery and annotation