Machine Learning Engineers

The use of machine learning algorithms allows companies, government agencies, and nonprofits—especially those that work with large amounts of data (known as Big Data) to gain insights (often in real-time), work more efficiently, and develop products and systems that would have been unattainable just a decade or so ago. Here are some examples of how machine learning is being used in different industries:

• banking and financial services: automated trading, fraud detection
• health care: to automate medical scan analysis, reduce the time it takes to make diagnoses and plan medical treatments, track the health of patients via wearable technology and the Internet of Things
• Internet and e-mail: online customer support, e-mail spam and malware filtering, content and product suggestions, customized newsfeeds, search engine result refining, online fraud detection
• languages: language translation services such as Google Translation
• oil and gas: to locate new energy resources, predict refinery sensor failure, make exploration and distribution more cost-effective
• transportation: self-driving vehicles, data analysis and modeling for delivery services and transportation planning, real-time monitoring and traffic control

Typical duties for machine learning engineers include designing and building algorithms and learning systems; developing processes and tools to monitor and assess model performance and data accuracy; generating new code and improving existing code; building data and machine learning platforms and pipelines; monitoring these systems to ensure that the software works correctly and logically; working with data scientists to analyze large, complex data sets in order to extract insights; providing support to product managers in implementing machine learning into a particular product; and collaborating with company executives and other IT professionals to develop new uses of machine learning in their company’s products and services.

Machine learning engineers may specialize in fields such as deep learning, computer vision, and natural language processing. Deep learning seeks to mirror the functioning of the human brain. In this subdiscipline, engineers create huge neural networks known as artificial neural networks that have at least three layers (but often more) of processing units in order to teach computers to recognize speech, identify images, and even make predictions. Each layer accepts and processes the information from the previous one as it learns a new task. In computer vision, huge neural networks are used to teach machines how to view and interpret the world around them by using data collected by cameras and other methods. Computer vision is being used is in the development of self-driving vehicles, smartphones that are programmed to unlock by scanning our faces, and scanning technology that has been found to be better at finding cancerous growths on mammograms than radiologists. Natural language processing (NLP) aims to teach computers to understand, interpret, and manipulate spoken and written human language.

Examples of NLP include e-mail filters, smart assistants, predictive text, language translation, and text analytics.