Machine Learning Engineer Intern - Scenario Generation

PlusAI•Santa Clara, CA

7d•$19 - $65

About The Position

PlusAI is a Physical AI company pioneering AI-based virtual driver software for factory-built autonomous trucks. Headquartered in Silicon Valley with operations in the United States and Europe, Plus was named by Fast Company as one of the World’s Most Innovative Companies. Partners including TRATON GROUP’s Scania, MAN, and International brands, Hyundai Motor Company, Iveco Group, Bosch, and DSV are working with Plus to accelerate the deployment of next-generation autonomous trucks. If you’re ready to make a huge impact and drive the future of autonomy, Plus is looking for talented individuals to join its fast-growing teams. The intern will explore and prototype a generative simulator that can synthesize diverse, realistic driving scenarios conditioned on map, traffic, and ego-vehicle state. The project will: Survey and evaluate existing open-source generative simulators / scenario generators for autonomous driving (diffusion based tools). Prototype a diffusion-based scenario generator that can take our internal dataset as input and generate new variations of scenes. Integrate the prototype into our internal data pipeline or simulation stack so that generated scenarios can be: exported as augmented training data, and/or used as test scenarios for downstream planning / prediction / control modules.

Requirements

Solid Python programming skills
Familiarity with deep learning frameworks (PyTorch preferred)
Basic understanding of generative models (e.g., VAEs, GANs, diffusion models)
Experience working with datasets (loading, preprocessing)

Nice To Haves

Prior exposure to diffusion models
Experience with autonomous driving datasets
Familiarity with 2D/3D coordinate systems, trajectories, and map representations
Experience with experiment tracking and ML tooling
Basic understanding of scenario-based testing for autonomous driving systems

Responsibilities

Evaluate existing tools for fidelity and efficiency to determine the best baseline.
Ingest and process internal, real-world driving logs to fine-tune the model.
Connect the prototype to the data pipeline, allowing it to seamlessly ingest map data and output structured scenarios.
Translate the generated scenarios into the standardized formats required by the company’s existing simulation engine.
Develop metrics to validate the physical realism, diversity, and kinematic feasibility of the generated scenarios.
Use generated scenarios to stress-test the vehicle's on-board modules against rare or safety-critical events.
Collaborate closely with ML researchers, simulation engineers, and on-board teams to define what makes a scenario useful for their specific needs.