Job Description
Job Detail
- Build and scale the training and inference systems that power Odyssey’s interactive video models, from large-scale distributed pipelines to real-time serving and optimization
- Work shoulder-to-shoulder with researchers to prototype new architectures, diagnose and improve model behavior, and bring ideas from paper to production
- Design high-performance data and compute systems for video generation and control, ensuring rapid iteration and efficient use of large-scale resources
- Develop tools, metrics, and visualizations that help the team understand how models behave and evolve over time
- Collaborate closely with product engineers to integrate Odyssey’s models into live, interactive user experiences that showcase new forms of generative media
- Move fast and iterate often
- You’ll be part of a small, focused team where experiments become demos, and demos become products
- Contribute to shaping Odyssey’s engineering culture: pragmatic, fast-moving, research-driven, and always oriented toward what’s possible next
-
Who You Are
A staff-level or senior engineer with experience in large-scale machine learning systems, distributed training, performance optimization, or model deployment - Deeply technical and hands-on: you find energy in writing code, optimizing kernels, and making systems faster, more reliable, and more elegant
- Motivated by bridging research and product
- You care as much about experimentation and insight as about performance and polish
- Comfortable operating in areas without precedent, creating tools, frameworks, and patterns as you go
- Thrives in small, collaborative teams where autonomy, clarity, and speed matter
- Passionate about building the next generation of generative and interactive media
- You are excited to help define how people create, play, and communicate through AI video
- ~