AI Systems Engineer – Vision & Agentic Evaluation

Apple Inc.; Sunnyvale (CA); Híbrido; USD 150.000 - 278.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Apple benefits

Descripción de la vacante

Apple Inc. is seeking an AI Engineer specializing in algorithm evaluation and agentic systems for advanced computer vision and video understanding.

You will lead benchmarking, integrate state-of-the-art models, and apply rigorous testing to ensure robustness and safety. As part of the DAQ team, you will design end-to-end evaluation pipelines, perform deep failure analysis, and architect autonomous multi-modal workflows that bridge experimentation with production.

Formación

• MS and a minimum of 3 years relevant industry experience.
• 3+ years of applied experience in ML, Computer Vision, or AI System Evaluation.
• Solid ML foundations: probability, statistics, data distributions, and bias/variance.
• CV and Vision-Language Models expertise; ViTs, temporal modeling, and video processing.
• Advanced evaluation skills: metrics, benchmarks, regression testing, and human-in-the-loop evaluation.
• Agentic systems: experience building and evaluating LLM/VLM-powered agents.
• Failure analysis: identify edge cases, hallucinations, and bottlenecks.
• Engineering excellence: Python and PyTorch for scalable evaluation pipelines.

Responsabilidades

• Design, build, and scale evaluation pipelines for image and video tasks.
• Perform deep failure analysis on model outputs, identifying root causes.
• Develop agentic workflows that autonomously solve multi-step problems.
• Lead data curation for high-quality, multi-modal benchmarks.
• Collaborate with core model training teams to share actionable metrics.

Conocimientos

Python

PyTorch

Computer Vision

Evaluation frameworks

LLM/VLM

Agentic Architecture

Educación

MS

Descripción del empleo

Apple Inc. is seeking an AI Engineer specializing in algorithm evaluation and agentic systems for advanced computer vision and video understanding.

You will lead benchmarking, integrate state-of-the-art models, and apply rigorous testing to ensure robustness and safety. As part of the DAQ team, you will design end-to-end evaluation pipelines, perform deep failure analysis, and architect autonomous multi-modal workflows that bridge experimentation with production.