Software Solutions Engineering Intern

Opens intel.wd1.myworkdayjobs.com in a new tab

This position focuses on the research and application of VLA (Vision-Language-Action) algorithms in robot scenarios, undertaking the whole process of VLA data collection, model fine-tuning, performance testing, problem analysis and optimization.
It requires proficient mastery of VLA algorithms (e.g., PI05) and the lerobot framework, as well as the ability to use Python to complete model training and optimization, so as to promote the efficient application of VLA technology in robot systems.

Be responsible for the collection, collation and verification of VLA (Vision-Language-Action) related data, ensuring data quality and completeness to meet the needs of model training and support model optimization in robot scenarios.
Perform fine-tuning on existing models, optimize the model training process in combination with the characteristics of VLA algorithms and robot application scenarios, and improve model adaptability and execution accuracy.
Design reasonable test plans to conduct comprehensive performance tests on the fine-tuned models, quantitatively evaluate key indicators such as model accuracy and generalization ability, and record and sort out test results in detail.
In-depth analyze problems arising during model testing, locate performance bottlenecks, and propose targeted optimization solutions combined with VLA algorithm principles and robot system characteristics for implementation.
Continuously track cutting-edge technologies in the VLA field and updates of the lerobot framework, iteratively optimize model performance, and ensure the stable and efficient operation of models in actual robot application scenarios.

Proficient in the principles and applications of VLA algorithms (e.g., PI05), and have an in-depth understanding of the core architecture of VLA models.
Familiar with the lerobot framework, understand its functions related to data collection and model deployment, and be able to use the framework to complete the implementation and debugging of robot-related algorithms.
Familiar with the basic principles and operation mechanism of robot systems, understand the basics of robot perception and control, and be able to optimize models in combination with robot scenarios.
Proficient in Python programming language, and familiar with common data processing and machine learning libraries (such as PyTorch, TensorFlow).
Have good problem analysis and solving abilities, strong team cooperation awareness, a serious and responsible work attitude, and a strong interest and research spirit in the field of VLA combined with robotics.

Sourced directly from Intel’s career page

Your application goes straight to Intel.

Similar Other roles