Heading

San Jose, CA

Internship

Onsite

$25/hr - $32/hr

Intern

Lyten is a supermaterial applications company that develops, sources, and manufactures low-carbon footprint products. They are seeking an ML/Data Science Intern to build and optimize time-series models from sensor data, collaborate with teams, and contribute to various data analysis tasks.

Apply Now

Responsibilities

Build, evaluate, and optimize time-series models for classification and event detection from multivariate sensor data.
Implement and experiment with models such as Random Forests, Gradient Boosting, LSTM, GRU, CNN-LSTM, and Temporal Convolutional Networks (TCNs).
Explore and experiment with emerging architectures such as Transformer-based time series models.
Apply windowing strategies (sliding/rolling windows) for preprocessing and feature extraction.
Evaluate models based on accuracy, precision, recall, latency, and inference speed.
Collaborate with firmware and backend teams to interface models with real-time sensor data pipelines.
Assist with developing low-latency ML inference modules for real-time deployment.
Perform exploratory data analysis and generate plots, model evaluation reports, and dashboards.
Contribute to internal documentation and (optionally) research publications.

Qualification

Required

Completion of high school and completion of sophomore level coursework in STEM course in University
Strong foundation in machine learning and deep learning principles.
Hands-on experience with time-series data, especially classification and sequence modeling.
Practical knowledge of Random Forests, LSTM, CNNs, or equivalent.
Proficient in Python and standard libraries (NumPy, Pandas, scikit-learn, TensorFlow or PyTorch).
Familiarity with data preprocessing techniques, including filtering and denoising.
Ability to write clean, reproducible, and well-documented code.
Pursuing or recently completed a Bachelor's/Master’s/PhD in Computer Science, Engineering, Physics, Applied Math, or related field.