data platform architect for physical AI

Grid Dynamics provides technology consulting, platform and product engineering, AI, and advanced analytics services. It helps enterprise companies undergoing business transformation address technical challenges and achieve positive business outcomes.

• Design data curation and annotation workflows from raw data ingestion to training dataset creation
• Define metadata registries and cataloging approaches for sensor assets, including machine and sensor identifiers, calibration versions, timestamps, pose, and data storage layers
• Establish data lineage, versioning, lifecycle management, and retention from raw sensor logs to training datasets
• Define how real-world data is prepared for simulation and how synthetic data is stored, versioned, and combined with real data
• Agree on data contracts with perception and sensor-kit teams
• Collaborate with the US-based Data Platform Architect Lead on the end-to-end data architecture
• Produce actionable architecture artifacts that engineering teams can implement
• Turn architectural decisions into working software quickly using AI coding tools

• Background in data architecture or engineering for large-scale platforms, with hands-on delivery
• Strong knowledge of unstructured and binary data, including images, video, and point clouds stored in object storage
• Hands-on expertise in metadata catalogs, registries, and asset management
• Solid understanding of ML data preparation, including curation, dataset splits, holdouts, and reproducible dataset versions
• Practical knowledge of data lineage and versioning from raw data to training datasets, using LakeFS, DVC, Delta, or equivalent tools
• Familiarity with lakehouse formats such as Parquet, Iceberg, or Delta
• Proficiency in batch and streaming ingestion using technologies such as Kafka or Kinesis
• Hands-on skills with at least one major cloud platform (AWS, Azure, or GCP) and data processing engines such as Spark or Databricks
• Proficiency in Python
• Ability to use AI coding tools to accelerate hands-on development and delivery
• Будет плюсом: background in autonomous vehicle, robotics, or ADAS data pipelines, familiarity with ROS bags, MCAP, sensor logs, drive data, or scene data, knowledge of annotation and labeling pipelines including auto-labeling, exposure to synthetic or simulated imagery and sensor data for training or testing, familiarity with ML experiment and artifact tracking tools such as MLflow

• Competitive salary
• Flexible schedule
• Benefits package including medical insurance and sports
• Corporate social events
• Professional development opportunities
• Well-equipped office