Imagry builds a generative AI platform (Cortex) for real-time autonomous driving using only camera vision, eliminating dependency on HD maps, LiDAR, or cloud connectivity. The tech stack—PyTorch, Python, C++, CUDA, and automotive-grade camera integration—reflects a compute-intensive, edge-first architecture designed for hardware-agnostic deployment. Engineering dominance in hiring (10 of 12 active roles) combined with active focus on perception algorithms, motion planning, and data pipelines signals heavy R&D investment in model training infrastructure and latency reduction, core technical challenges for production autonomy.
Imagry develops AI-driven autonomous vehicle technology for transit operators and automotive manufacturers. The platform runs on vision sensors alone, eliminating the cost and complexity of LiDAR and HD map dependencies, and operates without cloud reliance—critical for real-world deployment at scale. The company has deployed systems on public roads, including autonomous bus operations. Based in San Jose with teams across the United States, Israel, and Japan, Imagry operates as a privately held company founded in 2015 with 51–200 employees.
PyTorch, Python, C++, Linux, CUDA, OpenCL, and automotive Ethernet/CAN. The stack emphasizes GPU-accelerated deep learning and edge computing for real-time inference.
Primarily the United States, with active recruitment in Israel and Japan. Engineering roles represent 10 of 12 current open positions, skewed toward senior and mid-level engineers.
Fail-operational hardware systems, reducing inference latency, and ensuring sensor synchronization across camera arrays—all critical for production autonomous driving safety and real-time performance.
Other companies in the same industry, closest in size