Research Engineer

Oddin is a visionary company focused on revolutionizing digital content creation through AI. The Research Engineer role involves transforming research models into deployable Python services, collaborating closely with researchers to ensure seamless integration into the production platform.

Responsibilities:

• Package and containerize AI models (Python/Docker) from research into clean, versioned services with well-defined APIs
• Own the engineering side of the tech transfer process: inference specs, environment setup, model mocking, and integration scaffolding
• Collaborate closely with research scientists on quantization, TensorRT compilation, and hitting latency budgets (e.g. <200ms real-time response targets)
• Maintain and operate model services in production, debugging stability and performance issues under load
• Contribute to the dual-track delivery model — keeping the engineering platform moving even while research is still iterating

Requirements:

• Strong Python engineering skills; comfortable writing production-grade, maintainable code
• Comfortable owning ambiguous model-to-service transfers end-to-end with a high degree of autonomy
• Hands-on experience deploying ML/AI models (inference pipelines, serving frameworks)
• Familiarity with GPU workloads, containerization, and model optimization concepts
• Ability to read and work directly with research code and translate it into reliable services
• experience with video/image generation models
• TensorRT
• real-time streaming pipelines