AI/ML Engineer Intern

Description

Levisonics is a medical device startup dedicated to improving patient lives by developing innovative blood coagulation testing platforms that perform analysis with a single drop of blood using acoustic levitation. Our mission is to provide a safer and more convenient point of care platform for comprehensive assessment of bleeding and thrombosis, giving healthcare professionals accurate and reliable diagnostic tools to make informed decisions and improve patient outcomes.

We are seeking an AI/ML Engineer Intern to support development of machine learning models and signal processing methods used to interpret data from our device platform. Interns will work on translating raw acoustic and sensor signals into clinically meaningful coagulation measurements, gaining experience applying machine learning methods to a real medical device data set, alongside our software and biomedical engineering leads.

Responsibilities

1. Assist with development and evaluation of machine learning models for interpreting device signal data.
2. Support signal processing and feature extraction from raw sensor data.
3. Help validate model outputs against reference coagulation assays and clinical data.
4. Contribute to organizing datasets and pipelines used for model training and testing.
5. Document model architectures, experiments, and results.

Qualifications

1. Currently pursuing a degree in Computer Science, Data Science, Electrical Engineering, or a related field with an AI/ML focus.
2. Proficiency in Python and familiarity with machine learning libraries such as scikit-learn, PyTorch, or TensorFlow.
3. Familiarity with signal processing or time series data is a plus.
4. Interest in applying AI/ML methods to healthcare or medical device data.
5. Detail oriented, collaborative, and comfortable in a small startup team.

Learning Objectives

By the end of this internship, the intern will be able to:

1. Explain how machine learning models are applied to interpret raw sensor and acoustic signal data.
2. Build, train, and evaluate a model against a real validation data set with guidance from the team.
3. Connect model performance metrics to clinical measures of coagulation.
4. Document a machine learning experiment clearly enough for a teammate to reproduce it.

Contact Information