Bioscope AI
Bioscope AI is seeking an AI Engineer (Time-Series) to develop advanced multimodal AI systems that integrate time-series data from wearables and CGM with LLM-based reasoning. The role focuses on longitudinal patient analysis, bridging temporal and linguistic modalities, and deploying solutions that comply with privacy, security, and regulatory standards.
What You'll Do
Design and implement state-of-the-art multimodal AI models that combine time-series data from wearables, CGM, and other sensors with large language models for analysis and reasoning.
Build systems that analyze and interpret patient health trajectories, identifying patterns, anomalies, and clinically relevant insights from continuous monitoring data.
Create architectures that effectively merge sequential sensor data with natural language medical records, clinical notes, and knowledge bases.
Stay at the forefront of multimodal AI research, implementing and adapting the latest techniques in time-series forecasting, representation learning, and transformer-based models.
Collaborate closely with healthcare professionals to translate clinical needs into technical solutions.
Deploy robust, scalable AI systems that handle real‑world patient data while addressing privacy, security, and regulatory requirements.
Technical Focus Areas
Time‑series analysis and forecasting from wearable devices.
Integration of large language models with temporal biomedical data.
Multimodal representation learning and fusion techniques.
Anomaly detection and pattern recognition in longitudinal health data.
Temporal reasoning and causal inference from observational data.
Signal processing and feature extraction from continuous monitoring devices.
Qualifications Required
Master’s degree in Computer Science, Machine Learning, Biomedical Engineering, Statistics, or a related quantitative field (PhD preferred).
Strong foundation in deep learning frameworks, especially PyTorch.
Experience with foundation models and large language models (transformers, pre-training, fine-tuning).
Demonstrated experience with time-series analysis and forecasting methods.
Experience deploying deep learning models to production environments.
Proficiency in Python and modern ML development tools.
Strong understanding of machine learning fundamentals and statistics.
Ability to read and implement research papers.
Excellent communication skills and collaborative mindset.
Preferred
Publications in top-tier ML/AI conferences: NeurIPS, ICML, ICLR, KDD, etc.
Experience with healthcare data: EHR, wearables.
Background in signal processing or biosignal analysis.
Location: Boston, MA.
#J-18808-Ljbffr
What You'll Do
Design and implement state-of-the-art multimodal AI models that combine time-series data from wearables, CGM, and other sensors with large language models for analysis and reasoning.
Build systems that analyze and interpret patient health trajectories, identifying patterns, anomalies, and clinically relevant insights from continuous monitoring data.
Create architectures that effectively merge sequential sensor data with natural language medical records, clinical notes, and knowledge bases.
Stay at the forefront of multimodal AI research, implementing and adapting the latest techniques in time-series forecasting, representation learning, and transformer-based models.
Collaborate closely with healthcare professionals to translate clinical needs into technical solutions.
Deploy robust, scalable AI systems that handle real‑world patient data while addressing privacy, security, and regulatory requirements.
Technical Focus Areas
Time‑series analysis and forecasting from wearable devices.
Integration of large language models with temporal biomedical data.
Multimodal representation learning and fusion techniques.
Anomaly detection and pattern recognition in longitudinal health data.
Temporal reasoning and causal inference from observational data.
Signal processing and feature extraction from continuous monitoring devices.
Qualifications Required
Master’s degree in Computer Science, Machine Learning, Biomedical Engineering, Statistics, or a related quantitative field (PhD preferred).
Strong foundation in deep learning frameworks, especially PyTorch.
Experience with foundation models and large language models (transformers, pre-training, fine-tuning).
Demonstrated experience with time-series analysis and forecasting methods.
Experience deploying deep learning models to production environments.
Proficiency in Python and modern ML development tools.
Strong understanding of machine learning fundamentals and statistics.
Ability to read and implement research papers.
Excellent communication skills and collaborative mindset.
Preferred
Publications in top-tier ML/AI conferences: NeurIPS, ICML, ICLR, KDD, etc.
Experience with healthcare data: EHR, wearables.
Background in signal processing or biosignal analysis.
Location: Boston, MA.
#J-18808-Ljbffr