Staff AI/ML Engineer, Time-Series & Sensor Reasoning Models (Lorenz Labs)

As Staff AI Engineer, you design time-series foundation models that fuse multimodal sensor data to enable context-aware reasoning at the edge. You will drive research and architecture work within Lorenz Labs, influencing edge sensing solutions across automotive, health, and industrial domains. You tackle cross-modal alignment, embedding, forecasting, and anomaly detection to enable physically-aware AI. This role combines deep research with practical deployment, partnering with hardware and systems teams to deliver real-time, energy-efficient sensing. You’ll publish, present at top venues, and mentor junior researchers while shaping foundation-model strategy.

• medical, vision and dental coverage
• 401k
• paid vacation, holidays, and sick time
• discretionary performance-based bonus
• Lead R&D on intelligent time-series agents for edge, integrating multiple modalities (electrical, audio, motion, physiological) with text.
• Enable cross-modal use of additional contexts (text, image) in foundation models.
• Advance sensor fusion research for cross-domain alignment of diverse sensor domains.
• Create benchmarking pipelines for cross-domain time-series foundation models focusing on robustness, interpretability, and hardware performance.
• Apply parameter-efficient fine-tuning and alignment methods (LoRA, Q-LoRA, adapters, DPO, RLAIF).
• Push forward state-of-the-art time-series embeddings and compression for edge reasoning.
• Investigate modern foundation-model alignment techniques and co-design architectures with hardware, signal processing, and systems teams.
• Design statistical experiments for data collection, publish at ML and signal-processing venues, and mentor junior researchers.
• 6+ years of experience developing AI/ML products.
• Deep expertise in time-series ML, signal processing, and foundation models; hands-on training or fine-tuning of time-series foundation models.
• Proficiency in representation learning, time-series encoding, compression, and motif discovery in high-dimensional data.
• Knowledge of SOTA time-series reasoning models, cross-attention, multi-modal embedding, memory, and RAG.
• Experience with parameter-efficient fine-tuning (LoRA/Q-LoRA) and reward-based optimization methods (DPO, PPO, RLAIF).
• Strong statistical testing, experimental design, and causal discovery.
• Fluency in Python, PyTorch, and large-scale training pipelines on cloud or distributed systems (AWS, GCP).
• Ability to collaborate across ML, hardware, and embedded systems and deploy physical intelligence systems.
• cross-disciplinary collaboration
• mentoring and leadership
• communication of complex concepts
• time-series ML
• signal processing
• foundation models (Chronos, TimesFM, TimeGPT, etc.)