OverviewBe part of something innovative and cutting edge with a real-world impact. Join our team of more than two hundred software engineers, system architects, database engineers, graphic designers, project managers, QA analysts, and data analysts who are smart, creative, and excited by what they do. The American Institutes for Research (AIR) is a leading professional services firm with a growing software engineering and product development team. For nearly 70 years, we have designed, developed, and implemented large-scale assessment programs for states across the country with powerful results. AIR Assessment fuses statistics, technology, and content into powerful formative, summative, adaptive, alternate, and diagnostic assessments. These assessments help students learn, teachers teach, and parents know what is happening at school. Our years of work and experience in software design and development has positioned us as a leader in the online testing industry and provided us the opportunity to make a positive difference in the educational systems that impact millions of students across the nation. During the 2017-2018 school year alone, AIR delivered nearly 50 million online tests. Some of our ground-breaking work includes: ? Design and development of large-scale state assessment services ? Development of next-generation classroom products ? Advanced computer-adaptive algorithms (only one that's peer-approved in the country) ? Mobile support for the user interfaces ? Learning management systems with social media features ? User interfaces that are universally accessible to people with or without disabilities ? Innovative, machine-scorable items ? Business Intelligence and Machine Learning We are currently seeking Machine Learning Interns to join our Washington, DC office for the summer. The ideal candidate will start in May and work fulltime through mid-August. ResponsibilitiesWe are looking for a summer intern to join our machine learning team. This team focuses on the modelling of unconstrained text to support our educational assessment programs. The work will consist primarily of a well-defined project that helps us to improve our software. This project could include studies that: create and examine the impact of new features or models in scoring; examine elements of human rater scoring and their impact on modelling; examine the impact of sample size and other characteristics on modelling quality; or, examining the quality of various metrics on examining model quality. ? Implement and inform a project that is focused on improving some aspect of our machine learning software or methods. ? Write code in Python using machine learning frameworks and libraries to ingest and clean data, build models, and evaluate results ? Document, archive, and present work on the project ? Work as part of a multi-disciplinary team discussing methods and results of this project and others Qualifications? Level: Master's or Ph.D. degree in progress. ? Coursework in Computer Science, Mathematics, Computational Linguistics or similar field ? Experience with engine calibration methods (e.g., sampling, pipeline, analysis) ? Course and/or work-based experience with machine learning models including multi-layer neural nets, natural language processing (NLP) methods (e.g., text handling, grammars, corpora use, latent semantic analysis; entity recognition), and in the evalution of engine performance ? Coding ability with Python, machine learning frameworks (e.g., Keras, PyTorch) and related libraries for machine learning (e.g., scikit-learn) and NLP (NLTK, spaCy) ? Strong communication skills and ability to work with others ? An analytical mind with problem-solving abilities Associated topics: defect, qa, quality, quality assurance, quality check, quality control, quality manager, quality technician, requalification, test
* The salary listed in the header is an estimate based on salary data for similar jobs in the same area. Salary or compensation data found in the job description is accurate.