Open positions in Nokia Research Center, Palo Alto

The quantity and the complexity of the data is growing rapidly, as computing, sensing and storage are commodizing, the gap between physical and digital worlds in closing, and people are increasingly empowered to create, disseminate and consume information. Nokia is in the heart of such developments. Nokia produces in a year more than 500 million mobile devices with rich communication, media and sensing capabilities. In addition to the device manufacturing being significant logistics operations producing terabytes of data, Nokia's emerging services, software and content platforms result in various streams of diverse data.

We are starting a new team in Nokia Research Center Palo Alto, focusing on the challenges and opportunities of data. The new Data Insight team aims at performing world-class research on processes and techniques to derive insights to data. The team's approach combines the theoretical foundations of data analysis and computer science to hands-on skills on rapid development of scalable systems with practical relevance. The work includes gathering, analyzing, developing methods and building statistical and computational models from real-world datasets. The research topics include scalable mining of services and manufacturing, adaptivity in mobile terminals, and massively distributed information processing and collective mining in the planetary-scale networks of people and devices.

The Data Insight team is seeking researchers on data mining, machine learning, search engines, data visualization and scalable computing, ranging from fresh postdocs to most experienced research fellows. If you have excellent conceptual and theoretical skills in computer science and data analysis, interest to apply the skills in new contexts, strong problem solving skills and ability and urge to make impact beyond publications, promising results and toy examples, do not hesitate to contact us (email: Taneli.Mielikainen@nokia.com).

Data Mining and Machine Learning

As a data analysis expert you will be responsible for transferring terabytes of data into actionable models and descriptive knowledge. You are expected to have extensive expertise in machine learning and data mining both in theory and in practice. For instance, you can design and implement a kernel for a structured data sources and make the necessary approximations for scalability. You understand the merits and dismerits of probabilistic logic learning and know when it makes sense to use complex models. You see reasons why active learning is important for companies like Nokia. You have deep knowledge of different paradigms of data analysis, such as statistical learning theory, bayesian, algorithmic information theory, data mining, inductive inference, and so on, understanding the practical value and limitations of the paradigms. You are expected to challenge the paradigms and rethink even the foundations of data analysis as deemed necessary.

You are willing to use the techniques the problem requires, instead of hitting just with your favorite hammer or inventing a novel technique when not needed, and you are not allergic of even most mundane tasks when solving such task is relevant enough. You are motivated to take your research beyond data mining and machine learning, and creating usable prototype systems spurs you. You enjoy the art of programming and are able and keen in expressing your ideas fluently in variety of programming languages, such as C, Python, Haskell, Erlang and R. You can explain the difference between O(n), O(n log n), and O(n^2) in practice and know how to cope with computationally hard problems.

As you will working with an interdisciplinary team focusing on scalable solutions for data understanding and utilization, you should see the relevance of information visualization and scalable systems for data analysis research. You are enthusiastic to collaborate with scalability and visualization researchers in the team, eager to extend yourself to new topics and share your knowledge in data analysis with colleagues whose core competencies are in other topics.

Information Visualization and Representation

You will be responsible for transforming complex datasets and computational models to understandable and engaging visualizations and other representations. You are expected to push traditional limits of information visualization by producing unique representations of unique data sources together with the team. In many cases that requires also theoretical insights to mathematical properties of high-dimensional data spaces. The position requires ability to deeply understand multi-faceted and multi-dimensional information, as well as exceptional creativity that is required to condense and illustrate the most relevant aspects of the models and the underlying data.

The pivotal short-term role of the data visualization specialist in the team is to help communicating the results inside and outside the company, in the form of visual and often interactive representations, typically via the Web. As you are responsible for making the team's products engaging and visually appealing, you need to have solid sense of aesthetics and unparalleled skills in graphic design. As your visualizations will work as a communication medium and as a way to understand complex phenomena, you need to know how to convert seemingly chaotic information into an easily understandable and intuitive form. You are expected to be able and enthusiastic to implement even your most far-fetching ideas in practice, thus strong technical expertise with visualization tools and programming languages is a must. Relevant techniques include Processing, OpenGL, Flash and modern Web techniques such as Javascript, Ajax, and CSS. Expertise also on other programming languages, such as Python, Haskell, C and Matlab is highly desirable. As a member of small and agile team with a just-do-it attitude, you are ready to use your skills and creativity also in down-to-earth tasks, such as building web sites, demos and prototypes as needed.

You will be working closely with the team in understanding the semantics of the problem domain, defining and refining the analyses, and choosing the right tools for the tasks. Although you will be working closely with data mining and statistical modeling researchers, and no extreme knowledge in these areas is required, you should be well motivated to learn relevant techniques together with other team members. Likewise, willingness to educate colleagues on various visualization techniques and foundations is highly appreciated.

Scalable Computing and Data Management

You will be responsible for designing and implementing even the most demanding and experimental network setups, scalable data analysis platforms, high-performance clusters and storage solutions. You have deep understanding of distributed computing and scalable data management systems and methods (such as property testing and data stream algorithms) in practice and in theory knowing the current challenges and opportunities. You are willing to rethink the current distributed computing primitives and abstractions for data-driven systems, and assess their value also in practice.

The job includes maintenance and further development of the Data Insight Team's experimental server infrastructure, which serves as a company-wide example of a large-scale data analytics architecture. In this role, you will be consulting other teams and units in questions related to scalability and large-scale computing. Deep experience with Linux and other unices, as well as core internetworking technologies is a must. You know how to build a system that can handle 10K HTTP requests per second and achieves five-nines availability, or a terabyte-scale storage system that can store and access 10^9 data items efficiently without using a relational database. You can see beauty in simplicity.

You have a strong passion for programming, excellent coding skills and deep knowledge on programming paradigms and languages, such as Python, Erlang, C, Ocaml, and so on. Contributions to Open Source projects, especially in the field of scalable systems, form a big plus. You are able explain when it makes sense to parallelize even when the total work increased and when it is not worth the effort.

You will designing and implementing scalable systems for massive information processing with the team. On one hand, you have the ability and motivation to deepen your skills in data analysis and visualization methods with the team. On the other hand, you are motivated to get other team members on the map even if that slows down the progress in your own work for a while.