Senior Machine Learning Scientist (Poland Remote)

<h3>Company Description</h3><p>When you join Turnitin, you'll be welcomed into a company that is a recognized innovator in the global education space. For more than 20 years, Turnitin has partnered with educational institutions to promote honesty, consistency, and fairness across all subject areas and assessment types. Over 16,000 academic institutions, publishers, and corporations use our services: Gradescope by Turnitin, iThenticate, Turnitin Feedback Studio, Turnitin Originality, Turnitin Similarity, ExamSoft, Ouriginal and ProctorExam.</p><p>Turnitin has offices in the United States, Australia, India, Indonesia, Germany, Japan, Korea, Mexico, the Netherlands, the Philippines, Sweden, Ukraine, the United Kingdom. Our diverse community of colleagues are all unified by a shared desire to make a difference in education. Come join us, and let's make change together.</p><h3>Job Description</h3><p><strong>Responsibilities and Requirements</strong></p><p>We expect Senior Machine Learning Scientists to be versatile and have a well-balanced set of skills. You will focus on model training, with significant capacity for research (developing novel model architectures), dataset construction, and model hardening (preparing the model and code for production pipelines).</p><p>Day-to-day, your responsibilities are to:</p><ul><li>Work with subject matter experts and product owners to determine what questions should be asked and what questions can be answered.</li><li>Work with subject matter experts to curate, generate, and annotate data, and create optimal datasets following responsible data collection and model maintenance practices.</li><li>Answer questions and make trainable datasets from raw data, using efficient SQL queries and scripting languages, visualizing when necessary.</li><li>Develop and tune Machine Learning models, following best practices to select datasets, architectures, and model parameters.</li><li>Utilize, adopt, and fine-tune Language Models, including third-party LLMs (through prompt engineering and orchestration) and locally hosted LMs.</li><li>Stay current in the field - read research papers, experiment with new models and LLMs, and share your findings.</li><li>Optimize models for scaled production usage.</li><li>Communicate data insights, as well as the behavior and limitations of models, to peers, subject matter experts, and product owners.</li><li>Write clean, efficient, and modular code, with automated tests and appropriate documentation.</li><li>Stay up to date with technology, make good technological choices, and be able to explain them to the organization.</li></ul><h3>Qualifications</h3><p><strong>Required</strong></p><ul><li>Experience working with text data to build predictive models, both supervised and unsupervised.</li><li>A strong understanding of the math and statistics behind machine learning theory and fluency with general machine learning domains such as classification, regression, unsupervised clustering and recommender engines.</li><li>Software engineering background with 2-3 years of experience (we use Python, SQL, Unix-based systems, git, and github for collaboration and review).</li><li>Machine Learning development skills, including experiment tracking (we use AWS SageMaker, Hugging Face, transformers, PyTorch, scikit-learn, Jupyter, Weights &amp; Biases).</li><li>An understanding of Language Models, using and fine-tuning, encoding and decoding, and a familiarity with industry-standard LM families (such as BERT, GPT, and Bloom).</li><li>Bachelor’s or Master's degree in Computer Science, Statistics, Applied Mathematics or related field, with relevant industry experience, or outstanding previous achievements in this role.</li><li>Excellent communication and teamwork skills.</li><li>Fluent in written and spoken English.</li></ul><p><strong>Would be a plus</strong></p><ul><li>Familiarity in coding for at-scale production, ranging from best practices to building back-end API services or stand-alone libraries.</li><li>Essential dev-ops skills (we use Docker, AWS EC2/Batch/Lambda).</li><li>Familiarity in building front-ends (LLMs or more standard React, Javascript, Flask) for simple demos, POCs and prototypes.  </li><li>Experience with advanced prompting, fine-tuning or training an LLM, open-source or cloud, using industry accepted platforms (such as mosaic.ai or stochastic.ai).</li><li>Showcase previous work (e.g. via a website, presentation, open source code).</li></ul><h3>Additional Information</h3><p>Our Mission is to ensure the integrity of global education and meaningfully improve learning outcomes. </p><p><strong>Equality Statement</strong></p><p>Turnitin LLC and all related Group and Subsidiary Companies are committed to the policy that all persons have equal access to it programs, facilities and employment without regard to race, color, ancestry, national origin, age, gender, sexual orientation, gender identity, age, religion, creed, disability, medical condition, genetic