Industry Talks

We need your feedback! If you attend one of the Industry Talks, please take a moment to let us know about your experience using this form.

Session 1

Room X
Friday, 11:00–12:00


Hanna Pasula, Google

Hanna will discuss the issues involved in identifying matching video footage rapidly and at large scale. YouTube receives hundreds of hours of user-uploaded video every minute — and all of these uploads need to be compared against a huge library of copyrighted and abusive content. To do so quickly and accurately, we need to solve complex and interesting problems in video representation, index retrieval, and matching, all the while drawing on recent research and on Google's impressive computational resources. This talk will present some of our recent advances.

Hanna Pasula is currently an engineer at YouTube, where she is applying large neural networks and cutting-edge indexing algorithms to our enormous collection of videos. She received her PhD from UC Berkeley in 2003, and then worked as a postdoc at MIT and the University of Washington. Her research focused on probabilistic artificial intelligence; more specifically, on representations combining first-order logic with probability theory. In 2007, she joined Google Zurich, starting out in Search before settling down at YouTube.

Natural Language Processing Applications

Ketevan Tsereteli, Bloomberg

Automatically processing documents to uncover insights is extremely valuable in the low latency environment of the finance industry. The sensitive, multi-lingual, and high impact information in news, market, and economic data supplies a number of unique challenges. Bloomberg's NLP group tackles these issues through research in natural language processing and information extraction. This talk will focus on Named Entity Recognition and Disambiguation, including locating, classifying, and linking mentions of uniquely identified entities.

Ketevan Tsereteli is a research scientist and engineer from Bloomberg's R&D Knowledge Research group, where she works on foundational algorithmic problems across Artificial Intelligence. Keti's work spans semantic technologies for Bloomberg's Knowledge Platform, Natural Language Processing, and Question Answering.

Session 2: Room IV 13:30–14:30

Room X
Friday, 13:30–14:30

Cisco Collaboration- Be Inspired!

Elin Khosrawi Enmark, Cisco
Evelina Samuelsson, Cisco

What will Collaboration look like in the future? What kind of demands will future generation have on their Employer and the businesses they interact with? Cisco has a clear vision on how we will meet the demands of the future and change the way we work, live, play and learn with Cisco Collaboration technology.

Move Fast, Learn Things

Nishita Agarwal, Facebook

Nishita will cover the much talked about 'hacker culture' at Facebook. She will speak about how she's seen diversity play out in different situations at work, her attempts at failing, and a few tricks she’s learnt to handle situations better. Nishita will also talk about how Facebookers can borrow some interesting design philosophies from the products we build and instead apply them to ourselves in our day-to-day work.

Nishita Agarwal is a Software Engineer at Facebook, on the Friend Sharing team. She works on product experiences around sharing content on Facebook and loves to employ psychology in her work to create experiences people love to use. In the past, she has worked on Javascript infra, the Facebook mobile website and She has been at Facebook for over 2 years now, and recently moved from the Headquarters in sunny California to London to work on her British accent. When not working and thinking about new products, you can find her frequenting at improv shops, rocking Bollywood dance moves or binge-watching TV.

Session 3

Room X
Saturday, 11:00–12:00

Real-time audio processing and user driven development

Jon Bergenheim, Skype/Microsoft

Jon will present a typical VoIP audio pipeline and discuss some of the main difficulties each audio processing module is facing. He will spend the second half of the presentation on how Skype is leveraging its user base to drive and validate audio quality improvements.

Jon Bergenheim is a Principle Software Architect working as tech lead in the Skype audio team. He has been with Skype for 9 years and has 15 years of experience with real-time audio processing in VoIP clients. Jon is based in Uppsala where he also got his Master in Science degree in signal processing.

Google Anita Borg Memorial Scholarship, Engage Program and Community

Irina Koitz

Anita Borg was one of the first female computer scientists and her vision was that 50% of the computer science graduates would be women by 2020. Google is supporting her vision through the Google Anita Borg Memorial Scholarship, Engage Program and Community. This talk outlines the details on the Anita Borg programs, provides information on how to apply for the scholarship and highlights the engagement opportunities for scholars. Moreover, it gives an insider’s view of how it feels to be a Google scholar and showcases some of the achievements of the 2014-2015 community.

Irina Todoran Koitz is a PhD student in computer science at the University of Zurich, Switzerland, and was a Google Anita Borg Scholar in 2014. Irina received her BSc degree in computer science from the Polytechnic University of Bucharest, Romania, and the MSc degree with honors in service design and engineering from Aalto University (Helsinki University of Technology), Finland. Her interests include cloud requirements communication, understanding consumers through data-driven analysis, product management and empirical studies.