Description

Lead the design, pre-training, fine-tuning, and deployment of novel AI models for telematics applications.

This role is on-site.

Responsibilities

• Develop and implement novel algorithms representing physical phenomena from vehicle and human movements.
• Pioneer self-supervised learning techniques using multi-modal telematics sensor data.
• Build and manage scalable training and inference pipelines using Ray, PyTorch DDP, or Horovod.
• Integrate AI models into production systems while optimizing for cloud and edge deployment.
• Collaborate across engineering and product teams to translate research into product features.

Required Skills

• PhD or MS in AI, Computer Science, EE, Physics, Math, or Statistics.
• 7+ years of experience in AI/ML.
• 3+ years developing and deploying foundation models (e.g., BERT, GPT, custom transformers) for sequential/spatio-temporal data.
• Hands-on experience training multi-modal or time-series transformer architectures for sensor fusion.
• Deep expertise in designing pretext tasks for self-supervised learning on noisy sensor data.
• Proficiency in Python, Pandas, and NumPy.
• Extensive experience with PyTorch or TensorFlow for large-scale training.
• Experience with distributed training techniques and efficient large model methodologies.
• Experience building pipelines using Spark, Airflow, Docker, and cloud platforms (AWS, GCP, Azure).

Key Skills

Education

Bachelor's degree