Machine Learning Engineer - Multimodal Modeling

2 months ago

Base Salary

$250k - $295k/yr

Responsibilities

Design, train, and deploy multimodal machine learning systems spanning multimodal learning, physics-informed AI, digital twins, and spatial intelligence.
Own projects from problem definition and prototyping through production deployment, adoption, and ongoing performance monitoring.
Develop evaluation frameworks that compare model judgments with real business outcomes.
Build and extend scalable machine learning infrastructure, training-data pipelines, evaluation harnesses, and production monitoring.
Develop agentic model workflows with complex tool calls that interact with Stand’s world-modeling stack.
Build retrieval and similarity capabilities over learned representations of real-world assets.
Partner with the Platform team on model-harness interfaces and with Applied Science infrastructure engineers on production tooling.
Collaborate with technical subject matter experts and business teams across underwriting, pricing, mitigation, inspection, and customer decision-making.
Communicate technical decisions, tradeoffs, and project status while aligning work with business objectives.

Deep hands-on experience designing and training multimodal models that fuse heterogeneous data such as 3D or vision, simulation outputs, tabular data, and text.
Production experience training, evaluating, deploying, and iterating on multimodal models at scale.
Experience applying machine learning to complex physical systems, including domains such as atmospheric science, molecular modeling, protein modeling, robotics, or fluid dynamics.
Experience training or fine-tuning LLMs, including tool use, agentic workflows, or post-training methods.
Experience with retrieval and embedding systems, vector search, or similarity in latent space is preferred.
Experience with geometric deep learning, point clouds, meshes, and spatially aware architectures is preferred.
Familiarity with physics-informed AI and surrogate modeling across multiple domains is preferred.
Knowledge of geospatial, remote sensing, or Earth observation datasets is preferred.
Startup or zero-to-one technology development experience is preferred.
Strong project ownership, cross-disciplinary collaboration, communication, prioritization, and execution skills.
Candidates must be authorized to work in the United States; Stand does not sponsor new work visas, though TN, O-1A, and qualifying H-1B transfers may be considered.