Member of Technical Staff (Foundation Models)

Role overview Research and build foundation models and AI systems that learn from observations over time, forecast future outcomes, and support decisions under uncertainty. The work spans temporal and multimodal modeling, agentic forecasting, research methods, and the infrastructure needed to train and serve models at large scale.

Responsibilities - Design and train models on large, varied time-series and structured datasets, studying how data, model size, and compute affect performance. - Develop models that combine signals such as text, time series, tables, and events. - Build forecasting agents that retrieve information, use models and tools, reason about uncertainty, and synthesize results. - Develop efficient training and inference pipelines for low-latency, high-throughput forecasting. - Explore causal, counterfactual, and uncertainty-aware methods that connect predictions to decisions. - Adapt foundation models through pre-training, post-training, fine-tuning, and related approaches; evaluate results and move successful systems toward deployment.

Requirements - Strong experience across machine learning systems, from data and model design through training, evaluation, inference, and deployment. - Ability to work with large-scale or heterogeneous datasets and develop rigorous model evaluations. - Research judgment paired with the ability to build and ship working systems. - Experience with forecasting or temporal modeling, multimodal learning, or another relevant modeling area. - Ability to develop agentic workflows involving tools, retrieval, planning, or reasoning.

Nice to have - Training large foundation models from scratch, or experience with distributed training, GPU clusters, large data pipelines, and model optimization. - Work in time-series forecasting, causal learning, multimodal or generative modeling. - Experience with efficient inference, model serving, or agents combining language models, forecasting models, code execution, retrieval, and external tools. - Relevant research publications or substantial open-source contributions.

Benefits and work setup The position is full-time and remote, with the listed location covering the US and global candidates.