Project Domain
The project focuses on simulation data support for Autonomous Vehicles (AV) development, particularly for scenarios like obstacle detection, path planning, and complex traffic situations (e.g., unusual vehicles, traffic signs in tunnels, network interruptions). The candidate will work with sensor data collected from a test fleet of Avs, which include 8-12 cameras, LiDAR, and radar generating up to ~1TB of data per hour. A good understanding of AV context and real-world edge cases will be valuable.

Key Responsibilities

• Analyze large volumes of real-world sensor data to identify edge cases (e.g., hard braking events, nearby vehicles).
• Write complex SQL/Python/Spark (PySpark) queries to filter, aggregate, and transform time-series data.
• Work closely with internal tools (browser-based GUIs and backend services) for data search and auto-labeling workflows.
• Process structured and semi-structured data (e.g., array of structs from object detection models).
• Help identify data that will feed into AV simulation and ML training pipelines.
• Proactively propose and validate analytical approaches to improve data discovery processes.

Tech Stack

• Required: SQL (advanced), Python, Spark / PySpark, Data engineering skills
• Nice to have: Understanding of ML workflows