Overview
As a Software Engineer on the XR AI Systems team, you will architect and build distributed, asynchronous AI systems spanning smart glasses, mobile, and cloud. You will lead end-to-end integration of multimodal models into production, focusing on uncertainty-driven routing, calibration, abstention, and verification. You will optimize performance and reliability across device-to-cloud boundaries, balancing latency, bandwidth, energy, and cost. You will partner with ML researchers and hardware teams to design edge/mobile inference pipelines and scalable cloud serving, while delivering production observability and debugging tools. This role offers impact at scale, shaping the next generation of,
Responsibilities
- Architect and implement distributed, asynchronous AI systems spanning wearables, mobile, and cloud
- Lead end-to-end integration of computer vision, language, and multimodal models into production systems
- Design routing and decision mechanisms grounded in model uncertainty and verification
- Optimize cross-layer performance (latency, bandwidth, compute, memory, energy, battery life, cost)
- Co-design constrained edge/mobile inference pipelines with quantization and hardware acceleration
- Develop scalable cloud AI serving and service orchestration
- Establish evaluation frameworks, observability, and debugging tools to improve end-to-end performance and user experience
Key requirements
- Bachelor’s degree or equivalent practical experience
- 8 years of experience in C++ or Python and modern ML frameworks (e.g., JAX, PyTorch)
- Experience with on-device and cloud inference runtimes
- Proven ability to optimize latency, bandwidth, reliability, compute, memory, energy, and cost across device, mobile, and cloud layers
- Experience in asynchronous architectures, streaming data, state management, buffering, backpressure, fault tolerance, observability, and production debugging
- leadership
- cross-functional collaboration
- adaptability
- C++
- Python
- JAX
…
