About Throne

Throne is a continuous tracking device for getting personalized insight about gut health and hydration. Co-founded by John Capodilupo — co-founder and former CTO of WHOOP — we're bringing the rigor of continuous health tracking to two signals that have long been a guessing game. Our north star is to improve health and save lives.

AI / ML Engineer

We’re looking for an experienced AI / ML Engineer to own highly challenging sensing and machine-learning problems from definition through deployment. This is not a role where the roadmap will always arrive pre-scoped. You’ll be expected to understand product objectives, oversee the data and the broader eco-system from which it emerges, identify the highest-leverage technical work, make sound tradeoffs, and drive the solutions you craft through implementation, validation, deployment, and continuous improvement.

You’ll own model performance across computer vision, video, audio, and next-generation sensing systems, along with the data, evaluation, and production infrastructure required to improve that performance over time.

Location

Austin, Texas (In-person)

What You’ll Own

• End-to-end solution ownership means taking loosely defined product or sensing problems, determining what needs to be learned or built, identifying critical unknowns, developing a plan, executing it, and driving the work to a measurable outcome.
• Computer vision and video model performance across classification, segmentation, detection, embeddings, and temporal/video understanding.
• Large-scale dataset development, including data mining, labeling strategy, dataset quality, split design, active learning, failure analysis, and eval construction.
• Model training and pre-training, including fine-tuning, transfer learning, self-supervised or semi-supervised approaches. Pre-training base CV models experience is a strong plus.
• Model efficiency, including quantization, distillation, pruning, architecture selection, ONNX/TorchScript trade-off understanding, latency reduction, memory reduction, and cost/performance tradeoffs.
• Experimentation and evaluation rigor: metrics, ablations, cohort analysis, error taxonomies, regression tests, confidence calibration, and clear go/no-go criteria.
• Sensor validation for next-generation devices, using a scientific approach to compare sensors, characterize signal quality, design experiments, and determine whether new modalities improve real-world model performance.
• Data science and SQL workflows to inspect production data, evaluate model behavior, build analysis datasets, and connect model outcomes back to product and device behavior.
• Production ML integration, working with backend, firmware, and data systems to make models deployable, observable, reproducible, and reliable.
• Research translation, staying current with relevant CV, video, multimodal, compression, and foundation-model research and turning useful ideas into measurable improvements.

What We’re Looking For