Staff Time-Series & Sensor AI Architect

Analog Devices; San Jose (CA); On-site; USD 198,000 - 272,000; Full time

Benefits offered by this job

Medical coverage

Vision and dental coverage

401(k)

Paid vacation and sick time

Job summary

Analog Devices is seeking a Staff AI Engineer specializing in time-series and sensor foundation models. You will drive research that unifies multimodal sensor data into robust, context-aware systems for edge deployment, leveraging state-of-the-art time-series embeddings and cross-attention techniques.

You will collaborate across ML, hardware, and embedded teams to push forward real-time, energy-efficient sensing.

Qualifications

• Deep expertise in time-series ML, signal processing, and foundation models.
• Proficiency in representation learning, time series encoding, and motif discovery in high dimensional temporal data.
• Experience with parameter-efficient fine-tuning methods (LoRA, Q-LoRA) and reinforcement/optimization for multimodal data.
• Fluency in Python, PyTorch, and building large-scale training pipelines in cloud environments.

Responsibilities

• Lead R&D on intelligent time-series agents for edge computing by combining anomaly detection, reasoning, forecasting and foundation models.
• Integrate multiple data modalities such as electrical, audio, motion, and text to create cross-modal alignment.
• Create benchmarking pipelines for cross-domain time-series models, focusing on robustness, interpretability, and hardware performance.
• Publish and present work at venues like NeurIPS, ICML, ICASSP, KDD, and collaborate with hardware and systems teams.
• Mentor junior researchers and contribute to Lorenz Labs’ strategy for foundation models.

Skills

Time-series ML

Signal processing

Foundation models

Python

PyTorch

Multimodal embedding

LoRA/Q-LoRA

Education

Ph.D. in Electrical Engineering, Computer Science, or Applied Physics

Tools

Chronos

TimesFM

TimeGPT

Job description

Analog Devices is seeking a Staff AI Engineer specializing in time-series and sensor foundation models. You will drive research that unifies multimodal sensor data into robust, context-aware systems for edge deployment, leveraging state-of-the-art time-series embeddings and cross-attention techniques.

You will collaborate across ML, hardware, and embedded teams to push forward real-time, energy-efficient sensing.