Data Scientist

Glassbeam's mission is to enable smarter hospitals and labs with ML-powered products. Our products enable better medical device management, maintenance, and troubleshooting.

As a startup, we provide a positive work environment. No bureaucracy. No useless meetings. Clear goals and clarity of purpose. Spend your day building something useful and get satisfaction from work. We empower you to take ownership. You will be encouraged to come up with and try new ideas. Get a sense of achievement from delivering innovative products and playing an important role in the growth and success of a startup.

If you are an experienced ML practitioner who enjoys working in a startup environment and loves building ML-powered products that customers use daily, we would love to hear from you.

Role

We are looking for an ML practitioner with excellent data science and ML skills. The ideal candidate is highly skilled in:

• Framing business problems into ML/data science problems

• Navigating through uncharted waters and exploring complex messy unstructured data

• Applying data science/ML to create ML-powered products that customers use daily

Responsibilities

You will be applying data science and ML techniques on operational non-patient data obtained from medical devices to enable ML-powered products. A few examples of the business problems that these products will be solving:

• Enable predictive maintenance of medical devices

• Reduce unplanned downtime for medical devices

• Reduce mean time between failures of medical devices

• Detect early signs of deterioration in medical device health

• Enable proactive detection of a problem with a medical device

• Improve patient safety

• Enable better utilization of medical devices

• Improve operator productivity

• Improve device efficiency

You will own the complete end-to-end ML processes:

• Build domain expertise and deep understanding of different types of medical devices

• Formulate a business problem into a ML problem

• Conduct EDA on messy unstructured datasets to understand data and extract insights

• Create dashboards for showing insights extracted from data

• Identify data that can be used for ML from a complex and not-well defined datasets

• Determine the different ML techniques that can applied on the operational data collected from medical devices

• ETL, cleanse and prepare data for different ML use cases

• Apply ML algorithms, both supervised and unsupervised

• Train and evaluate ML models

• Collaborate with others to deploy, monitor, and manage models in production

• Manage model lifecycle

Requirements

Minimum

• 5+ years' work experience as a data scientist training and deploying ML models in production

• 3+ years' work experience training and deploying predictive maintenance models in production

• 2+ years' work experience training and deploying anomaly detection models in production

• 2+ years' work experience using Python for EDA, data prep, feature engineering, and ML

• 2+ years' work experience with SQL

• Deep understanding of different ML algorithms, frameworks, and libraries

• Strong knowledge of statistics

• Excellent exploratory data analysis skills

• Strong data cleansing, data prep and feature engineering skills

• Experience working with big data technologies such as Spark and HDFS

• Master’s or Ph.D. in Math, Statistics, CS, or other quantitative fields

Plus

• Experience with MLFlow, KubeFlow, AirFlow

• Experience with Spark Mllib

• Experience programming in Scala

• Experience with HDFS, Cassandra, Vertica,

• Experience with Tableau, Superset