SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
SpAItial is building the next generation of World Models—generative AI systems that natively understand the physics and geometry of 3D environments. The company is applying this technology across robotics, AR/VR, gaming, and cinema.
As an ML Inference Engineer, you will own the backend systems that serve SpAItial's world models in production. You'll be responsible for building and operating the inference APIs that power the product, transforming frontier research models into fast, reliable, and scalable services. This is a hands-on role requiring deep experience shipping production AI systems.
Key responsibilities:
- Own the inference services and APIs from request to generated output
- Optimize serving performance, reliability, and cost at scale, maximizing GPU utilization and minimizing device memory footprint
- Apply latest techniques for improving inference performance (profiling, memory optimization, batching, cold start reduction)
- Deploy research models as production services, including customer-specific configurations
- Collaborate with research and product teams to bring new models to users
This role sits at the intersection of research and production engineering. You'll work closely with researchers to understand model requirements and with product teams to ensure reliability and performance meet user needs.
Requirements:
- 3+ years of software engineering in an AI, machine learning, or computer vision product environment
- Proven experience designing and owning production APIs
- Strong Python skills and hands-on experience shipping containerized services on GPU inference platforms
- Demonstrated expertise in GPU inference performance optimization: profiling, memory management, batching, and cold start optimization
- Familiarity with 3D or computer vision data is a strong plus
