Over the past three years, Philip Morris International (PMI) has made a dramatic pivot away from the marketing and sales of combustible cigarettes, and moved towards its Reduced-Risk Product (RRP) portfolio, focusing on its flagship product, IQOS. This launch of this product is PMI’s step forward to creating a smoke-free future for legal age consumers who continue to enjoy tobacco, but are seeking smoke-free alternatives. All of this has been accompanied by an organization-wide transformation in ways of working, culture, and strategy, with emphasis on digital, consumer and collaborator strategies.
Do you look for ways to learn and grow in an international environment? Do you want to build a remarkable career? Are you up for a 6-month challenge starting as of July 2019? Then apply now, we need you:
Intern in Natural Language Processing
We are looking for a Machine Learning Intern to join PMI Data Science Team. If you have experience in natural language processing and are passionate about building machine learning systems to support the insights of human experts we would love to hear from you.
PMI undergoes a major transformation on its journey towards a Smoke Free Future. The success of this transformation relies on involvement and engagement of PMI employees. Hence, PMI carries on an ongoing conversation with its employees and leverages the state-of-the-art technology to uncover powerful insights from employee feedback and understand what people really mean.
Continuous listening strategy allows collecting a lot of feedback, however its value is only unlocked by the resulting meaningful actions. As an intern, you will assess currently used solutions and develop custom employee feedback classifiers aiming to maximize the support for insights and actions.
• Use natural language processing (NLP) and machine learning techniques to analyze employee feedback (sentiment, attitude, intent, subject)
• Determine the most relevant features that drive insight into the data
• Develop an approach to address labelled data bottle-neck (transfer learning, weak supervision)
• Compare the performance of general domain, broad business domain and specific domain corpora
• Evaluate the effect of corpus specificity and size
• Assess the quality of support provided by currently used and novel solutions for insights and actions
Skills and qualifications:
• Currently studying for your master degree in Computer science or related area
• Understanding of NLP concepts, such as word embeddings, PoS tagging
• Familiarity with ML frameworks (such as Pandas, Numpy, Scikit-learn, Pytorch)
• Solid programming skills in Python
• A desire to learn and strong motivation to succeed
• Unix shell scripting
• Knowledge of neural network architectures for text (RNN, CNN, LSTM)
Philip Morris International: Building a Smoke-Free Future
Philip Morris International (PMI) is leading a transformation in the tobacco industry to create a smoke-free future and ultimately replace cigarettes with smoke-free products to the benefit of adults who would otherwise continue to smoke, society, the company and its shareholders. PMI is a leading international tobacco company engaged in the manufacture and sale of cigarettes, smoke-free products and associated electronic devices and accessories, and other nicotine-containing products in markets outside the U.S. PMI is building a future on a new category of smoke-free products that, while not risk-free, are a much better choice than continuing to smoke. Through multidisciplinary capabilities in product development, state-of-the-art facilities and scientific substantiation, PMI aims to ensure that its smoke-free products meet adult consumer preferences and rigorous regulatory requirements. PMI's smoke-free IQOS product portfolio includes heated tobacco and nicotine-containing vapor products. As of Dec. 31, 2018, PMI estimates that approximately 6.6 million adult smokers around the world have already stopped smoking and switched to PMI’s heated tobacco product, which is currently available for sale in 44 markets in key cities or nationwide under the IQOS brand. For more information, see our PMI and PMIScience websites.