Lead AI Architect (AIOS)

About the role

As a world‑leading OEM, Mercedes‑Benz is seeking a self‑motivated Lead AI Architect to design, develop and evolve an end‑to‑end vehicle‑cloud integrated AI architecture across major vehicle domains.

Key responsibilities

• Own the end‑to‑end vehicle‑cloud integrated AI architecture, covering cloud AI platform, on‑board deployment, cross‑domain orchestration and model lifecycle management.
• Drive AI use‑case implementation from requirements and design through model selection, training, fine‑tuning, vehicle‑side optimization, integration and mass‑production release.
• Design and optimise cloud‑side AI capabilities, including model evaluation, training data preparation, fleet‑based updates and collaborative inference.
• Lead on‑board AI model adaptation, handling quantisation, latency, memory and power/thermal optimisation for automotive SoCs and domain controllers.
• Translate research results, open‑source models and prototypes into stable, automotive‑grade AI functions that perform under real‑world conditions.
• Collaborate closely with cloud platform, hardware, software and testing teams to align architecture, resolve blockers and accelerate feature delivery.

Required profile

• 5+ years of hands‑on, production‑oriented AI engineering experience, preferably in automotive or embedded systems.
• Strong expertise in model training, fine‑tuning, optimisation, quantisation and deployment.
• Deep understanding of vehicle‑cloud integrated AI systems, including training pipelines, model management, OTA updates and closed‑loop data.
• Knowledge of multiple vehicle domains (cockpit, ADAS/ADS, body, chassis, powertrain) and automotive E/E architecture.
• Proven experience in on‑vehicle integration, debugging and delivering AI features from prototype to mass production.
• Excellent cross‑team communication, stakeholder alignment and execution skills.
• Strong analytical, problem‑solving and technical leadership abilities.
• Fluent English, both written and spoken.

Required skills

• Computer vision
• Large language models
• Model training and fine‑tuning
• Model optimisation and quantisation
• Deployment on automotive SoCs and domain controllers
• Cloud AI platform development
• OTA model updates
• Closed‑loop data and model lifecycle management