Learn about the updates being made to TensorFlow in its 2.0 version. We’ll give an overview of what’s available in the new version as well as do a deep dive into an example using its central high-level API, Keras. You’ll walk away with a better understanding of how you can get started building machine learning models in Python with TensorFlow 2.0 as well as the other exciting available features! EVENT: EuroPython 2019 SPEAKER: Brad Miro PUBLICATION PERMISSIONS: Original video was published with the Creative Commons Attribution license (reuse allowed) ATTRIBUTION CREDITS: Original video source: https://www.youtube.com/watch?v=9CdCczxwuZ4 https://www.youtube.com/watch?v=qxGxmU58HJc

If toasters talking to fridges is no joke to you, then you are aware of the big Internet of Things hype these days. While all kind of devices get connected and hacked, one of the oldest class of IoT devices seems to be forgotten even though it is literally everywhere – VoIP phones. For configuration and management purposes, VoIP phones run a web application locally on the device. We found several critical bugs (reported CVEs) in the web application as well as in the webserver which enabled us to hijack the phones. Starting with simple XSS and CSRF issues, via command injections and memory corruptions right through to remote code executions, all popular vulnerability classes can be found on those devices. We will present our findings together with the tools and strategies we used, and will enable you to do the same with your own phones and other IoT devices. Further, we will provide helpful ARM shell code patterns, scripts and tricks which hackers can use to find bugs. We will conclude our talk by showing that automatic tools fail to discover such vulnerabilities. Therefore, manual IoT pentesting is still required. If you think these management interfaces are not exposed to the internet, you are wrong. In a scan, we found thousands of reachable phones vulnerable to our exploits. EVENT: DEF CON 27 SPEAKER: Stephan Huber PUBLICATION PERMISSIONS: Original video was published with the Creative Commons Attribution license (reuse allowed). ATTRIBUTION CREDITS: Original video source: https://www.youtube.com/watch?v=4Ing58vWkx0 https://www.youtube.com/watch?v=sXiGknzD02c

At Twiggle, we learn and understand users' queries and match their relevant products. To that end, we are building our own ontology of the world, and an advanced Natural Language Analyzer which aims to map both queries and products from across the e-commerce domain onto our ontology. Building an ontology is a tedious task which requires both specialized expertise as well as a broad look on the way the world is built. To facilitate that process, we have developed a unique algorithm. The algorithm generates a data-driven ontology by applying Node2Vec and clustering methods on query-to-product clicks along with minimal information about these products. In this talk, we will discuss this algorithm and offer a deep dive into the methods we used. EVENT: PyData Tel Aviv 2019 SPEAKER: Ella Bor PUBLICATION PERMISSIONS: PyData provided Coding Tech with the permission to republish PyData talks. CREDITS: PyData YouTube channel: https://www.youtube.com/channel/UCOjD18EJYcsBog4IozkF_7w https://www.youtube.com/watch?v=QnMqw0zL7ns