Apple
AI Engineer – Algorithm Evaluation & Agentic Systems
$150K–$278K
Sunnyvale, California · Full-time · Posted Aug 12, 2026
- Experience
- Minimum 3 years relevant industry experience; 3+ years of applied experience in Machine Learning, Computer Vision, or AI System Evaluation.
- Education
- MS
About the job
Evaluates and benchmarks image and video understanding models, analyzes their failures, and builds agentic workflows for multimodal tasks.
What you'll do
- Design and scale end-to-end and component-level evaluation pipelines and benchmarks.
- Analyze model failures, including visual hallucinations, temporal inconsistencies, and edge cases; provide actionable findings to model training teams.
- Build, deploy, and evaluate agentic workflows, and lead curation of datasets and ground-truth benchmarks.
What they ask for
- MS degree.
- At least 3 years of relevant industry experience and 3+ years of applied experience in Machine Learning, Computer Vision, or AI System Evaluation.
- Deep understanding of Computer Vision and Vision-Language Models, including Vision Transformers, spatial-temporal modeling, and image/video processing.
- Experience defining evaluation metrics and frameworks, evaluating agents, analyzing failures, and proficiency in Python and PyTorch.
Nice to have
- Experience leading evaluation strategies and mentoring engineers; strong foundation in statistics and experimental design.
- Knowledge of reinforcement learning, planning, or decision-making systems; experience with multimodal or multi-agent systems, AI reliability, safety, or benchmarking.
Pay
$150,400 to $277,600 USD base salary
Tools
Python, PyTorch, Vision Transformers (ViTs), LLM/VLM-as-a-judge
AI focus
Robust evaluation and safety of computer vision and video understanding models; agentic multimodal workflows; failure analysis and benchmarking.
Model-written summary · the original posting is authoritative