Staff Machine Learning Engineer

Workman Labs; Mountain View (CA); On-site; USD 218,000 - 328,000; Full time

Job summary

Unity is seeking a Staff ML Engineer to lead computer vision and multi-modal AI initiatives, taking models from research to production for billions of players.

The role requires deep CV and multi-modal expertise, strong Python/PyTorch skills, and a proven leadership track record in delivering scalable ML systems in on-site Mountain View, CA.

Qualifications

• 6+ years ML engineering with depth in CV and/or multi-modal modeling.
• Production experience with transformer-based and diffusion vision models (ViT, CLIP, Stable Diffusion, DETR, SAM).
• Strong full model lifecycle knowledge: data curation, training, evaluation, serving at scale.

Responsibilities

• Establish technical vision and roadmap for CV and multi-modal AI.
• Drive design of systems for image/video understanding, generation, segmentation, detection, and multi-modal reasoning.
• Make architectural decisions balancing quality, latency, and cost across cloud, server, and on-device targets.
• Own the full pipeline from research prototypes to production: training, fine-tuning, distillation, serving.
• Lead and mentor ML engineers; define and track KPIs for model quality, accuracy, latency, memory, and cost.

Skills

Computer vision

Multimodal modeling

Transformer models

Diffusion models

Python

PyTorch

Leadership

Model deployment

Tools

CoreML

TFLite

ONNX

Job description

Unity is building next-generation AI-driven gaming experiences with generative world models, neural rendering, and multi-modal understanding — this hands-on technical leadership role brings cutting-edge computer vision and multi-modal models into production for billions of players.

$218,400 – $327,600 / yr

Location

Mountain View, CA

On-site

Seniority

Staff

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

What you'll do

Responsibilities

• Establish technical vision and roadmap for computer vision and multi-modal AI (transformers, diffusion models, vision-language models)
• Drive design of systems for image/video understanding, generation, segmentation, detection, and multi-modal reasoning
• Make architectural decisions balancing quality, latency, and cost across cloud, server, and on-device targets
• Own the full pipeline from research prototypes to production: training, fine-tuning, distillation, serving
• Apply compression, quantization, pruning, and knowledge distillation for constrained environments Unique
• Lead and mentor ML engineers; define and track KPIs for model quality, accuracy, latency, memory, and cost

What you'll need

Experience

• 6+ years ML engineering with significant depth in computer vision and/or multi-modal modeling
• Production experience with transformer-based and diffusion vision models (ViT, CLIP, Stable Diffusion, DETR, SAM)
• Strong full model lifecycle knowledge: data curation, training, evaluation, serving at scale
• Familiarity with efficient attention, diffusion samplers, multi-modal fusion, and vision-language alignment Unique
• Strong Python and PyTorch skills; proven technical leadership track record
• Preferred: world-model/video-generation/neural rendering pipelines (NeRF, 3DGS), on-device deployment (CoreML, TFLite, ONNX), real-time graphics or game engine background Common