ML Engineer for Generative Video Systems (Low Latency)

Mirage; New York (NY); On-site; USD 140,000 - 200,000; Full time

Benefits offered by this job

Medical, dental, vision plans

401K match

Commuter benefits

Catered lunches

Job summary

Mirage is hiring an ML Engineer in NYC to build and scale video generation models powering Captions. You will work at the intersection of research and systems engineering to accelerate diffusion and autoregressive generation with production-ready tooling.

Join a team focusing on faster, more efficient models with ultra-low latency, real-time generation, and scalable distributed training infrastructure. In-person work at our NYC HQ is often required.

Qualifications

• 2+ years of professional industry experience.
• Experience with deep learning systems and infrastructure.
• Ability to move quickly from prototype to production.
• Strong debugging and performance profiling skills.

Responsibilities

• Train and optimize large-scale video and multimodal models.
• Improve efficiency across training and inference (memory, latency, cost).
• Implement distillation, quantization, and pruning to accelerate generation.
• Build and maintain distributed training systems.
• Optimize GPU utilization, parallelism, and throughput.
• Develop tooling for experimentation, evaluation, and debugging.
• Translate research models into production-ready systems.
• Monitor and improve model performance in real-world usage.

Skills

Deep learning systems

Performance profiling

Prototype to production

Low-latency inference

Education

BS/MS/PhD in CS/ML or related

Tools

PyTorch

CUDA

Triton

Distributed training

Job description

Mirage is hiring an ML Engineer in NYC to build and scale video generation models powering Captions. You will work at the intersection of research and systems engineering to accelerate diffusion and autoregressive generation with production-ready tooling.

Join a team focusing on faster, more efficient models with ultra-low latency, real-time generation, and scalable distributed training infrastructure. In-person work at our NYC HQ is often required.