Inference Specialist, Creative Technology - InterPositive

Workman Labs; Los Angeles, Northern (CA, KY); Hybrid; USD 165,000 - 265,000; Full time

Job summary

Netflix is seeking an Inference Specialist on its Creative Technology (InterPositive) team to run and debug custom generative-AI inference workflows for film and series, operating GPU inference jobs, tuning parameters for production-quality results, and building reliable, reproducible pipelines alongside researchers and artists.

The role requires hands-on experience with PyTorch/CUDA, Python-based ML code, and Linux, plus strong collaboration across technical and creative teams to ensure

Qualifications

• 4+ years in ML production, VFX technology, post-production engineering, or related technical roles.
• Hands-on experience running GPU-based model inference for generative AI systems.
• Python-based ML codebases and Linux command-line expertise.
• Experience debugging production runs using logs and configuration files.
• Working knowledge of deep learning inference concepts.
• Experience with video/image production formats (ProRes, H.264/H.265, EXR, PNG, MP4/MOV).
• Ability to coordinate across technical and creative stakeholders.

Responsibilities

• Operate custom generative AI inference workflows across film and series projects.
• Run, monitor, and troubleshoot GPU-based inference jobs across workstations, cloud, and cluster environments.
• Prepare and validate inputs for model inference (video, image, audio, masks, conditioning assets).
• Tune inference parameters to achieve production-quality results.
• Debug failed runs by inspecting logs, outputs, configs, and model checkpoints.
• Maintain clean, repeatable inference launch workflows with scripts and documentation.
• Partner with researchers to test new models in production scenarios.
• Support rapid iteration with artists and creative stakeholders.
• Bridge communication between creative, production, research, and engineering teams.

Job description

Inference Specialist, Creative Technology - InterPositive

Netflix is hiring an Inference Specialist on its Creative Technology (InterPositive) team to run and debug custom generative-AI inference workflows for film and series — operating GPU inference jobs, tuning parameters for production-quality results, and building reliable, reproducible pipelines alongside researchers and artists.

$165,000 – $265,000 / yr

Location

—

Seniority

—

Compared with 41 other Engineering roles: Common across similar roles Unique to this role

What you'll do

Responsibilities

• Operate custom generative AI inference workflows across film and series projects
• Run, monitor, and troubleshoot GPU-based inference jobs across workstations, cloud, and cluster environments
• Prepare and validate inputs for model inference (video, image, audio, masks, conditioning assets)
• Tune inference parameters to achieve production-quality results
• Debug failed runs by inspecting logs, outputs, configs, and model checkpoints
• Maintain clean, repeatable inference launch workflows with scripts and documentation
• Partner with researchers to test new models in production scenarios
• Support rapid iteration with artists and creative stakeholders
• Own quality control for generated outputs Unique
• Bridge communication between creative, production, research, and engineering teams

What you'll need

Experience

• 4+ years in ML production, VFX technology, post-production engineering, or related technical roles
• Hands-on experience running GPU-based model inference for generative AI systems Common
• Python-based ML codebases and Linux command-line expertise
• Experience debugging production runs using logs and configuration files Unique
• Working knowledge of deep learning inference concepts Unique
• Experience with video/image production formats (ProRes, H.264/H.265, EXR, PNG, MP4/MOV)
• Ability to coordinate across technical and creative stakeholders
• Effectiveness in fast-moving R&D environments Unique
• Tools: PyTorch, CUDA, ffmpeg, OpenCV, NumPy, safetensors, Git; Slurm/job scheduler experience preferred Unique