Multimodal ML Engineer for Physics‑Driven Risk Modeling

Stand; San Francisco (CA); On-site; USD 250,000 - 295,000; Full time

Benefits offered by this job

Health insurance

Dental & Vision plans

Weekly meals or stipends

401(k) plan

Job summary

Stand is seeking a Machine Learning Engineer for the Applied Science team to design, train, and deploy flagship AI capabilities. You will work on multimodal meshing of the Stand World Model with language models, integrating physical simulation, 3D asset representations, and business context for improved underwriting, pricing, and mitigation.

You will own modeling end‑to‑end, from architecture and training strategy through evaluation and production deployment, partnering with Platform to ensure

Qualifications

• Deep experience designing and training multimodal models (3D/vision, simulation, tabular, text).
• Proven production experience: training at scale, evaluation, deployment, iteration.
• Experience applying ML to complex physical systems (various domains).
• Experience training or fine-tuning LLMs with tool use and agentic workflows.
• Strong ownership: planning, prioritization, delivery of complex work.
• Ability to connect technical work to business objectives.

Responsibilities

• Design, build, and deploy ML systems for multimodal learning, physics‑informed AI, and spatial intelligence.
• Own projects end‑to‑end from problem definition to deployment and monitoring.
• Develop evaluation frameworks comparing model judgments with business outcomes.
• Extend scalable ML infrastructure and tooling.
• Collaborate with Platform team on model harness interfaces.
• Communicate decisions, tradeoffs, and status across teams.

Skills

Multimodal models

Production ML

Physics-informed ML

LLM fine-tuning

Project ownership

Cross-disciplinary

Strong communication

Fast-paced environments

Job description

Stand is seeking a Machine Learning Engineer for the Applied Science team to design, train, and deploy flagship AI capabilities. You will work on multimodal meshing of the Stand World Model with language models, integrating physical simulation, 3D asset representations, and business context for improved underwriting, pricing, and mitigation.

You will own modeling end‑to‑end, from architecture and training strategy through evaluation and production deployment, partnering with Platform to ensure