AI Research Scientist- Multimodal Foundational Models

Bosch Group•Sunnyvale, CA

About The Position

The Bosch Research and Technology Center North America conducts research in areas including artificial intelligence, energy technologies, internet technologies, circuit design, semiconductors, wireless, and advanced MEMS design. Its Foundation Model Powered AI Enablers group researches and develops scalable and transparent AI and big-data analytics solutions using data such as audio, images, and sensor logs for domains including Industry 4.0, IoT, autonomous driving, and connected vehicles. The group collaborates with academic and industry organizations and publishes research in leading AI conferences and journals. The AI Research Scientist will develop and lead research projects involving multimodal sensor data and foundation models, with applications in vehicle ADAS and other Bosch products. The position includes designing and validating multimodal deep-learning and time-series foundation models, extending those models with information from image and video modalities, advising management on relevant technologies, and sharing research through publications or patents.

Requirements

Ph.D. in Computer Science, Electrical Engineering, Information Technology, or a related discipline, or a master's degree with 2–3 years of preferred professional experience.
Expertise with time-series foundation models beyond forecasting tasks.
In-depth experience in signal processing for sensor data and integrating sensor data with deep-learning methods.
Proficiency in Python and PyTorch, including libraries such as torchaudio, torchvision, and torchmetrics; familiarity with PyTorch Lightning.
Strong publication record in relevant venues such as ICASSP, NeurIPS, Interspeech, ICML, ICLR, KDD, ICRA, CVPR, ICCV, and ECCV, or equivalent contributions such as patents or significant open-source projects.
Strong interpersonal, communication, and teamwork capabilities.

Nice To Haves

Three or more years of experience in industrial research.
Experience in one or more of data-centric AI, synthetic data generation, or agentic AI.
Proficiency with Git and an integrated development environment such as VSCode or PyCharm, and experience with experiment-tracking tools such as MLFlow.
Familiarity with high-performance computing systems and job schedulers such as Slurm or LSF.
Hands-on experience in product development in the stated areas for consumer or enterprise markets.
Experience leading projects with small teams, mentoring junior researchers and interns, managing project timelines, and delivering results within time constraints.

Responsibilities

Develop and lead AI research projects that label, predict, classify, cluster, describe, and fuse multisensor data, including acoustic, telemetry time-series, vibration, radar, lidar, image, Wi-Fi, and ultrasound data, to improve or enable ADAS functionality in vehicles and AI capabilities in other Bosch products.
Architect, design, and validate multimodal deep-learning and time-series foundation models for multivariate time-series signals.
Integrate and extend time-series models to use auxiliary modalities such as video and images to improve context understanding.
Provide expert insights to management to support strategic planning, R&D direction, and investment decisions in relevant technology sectors.
Keep up with technological innovations and document and disseminate research findings through high-caliber publications and/or patent submissions.