I am a business leader committed to solving impactful problems at scale with a track record of bringing strategic vision and global cross-functional programs to life.In addition, I possess a vast international and management experience combined with a strong passion about developing solutions for large-scale problems, ranging from organizational transformation to specific problems, such as digital transformation in various industries.For Google, I work cross-functionally with product, engineering and developer relations teams in order to bring large-scale open source strategies to life. I have led strategic partnerships and implemented global community building strategies for several technologies, such as Beam, Airflow, Spark, Hadoop, Kubernetes, and open source products, such as CDAP, Dataproc and Dataflow.Having worked as a Senior Program Manager for Fortune 500 clients, the United Nations, World Health Organization, UNESCO and governments, I thrive in challenging environments which require strategic planning and strong project leadership to succeed.Moreover, I speak 5 languages and I enjoy bringing a multi-cultural approach to everything I do.I hold a Master in Pharmaceutical Sciences and graduated from the IE Executive MBA with honors on both the final pitch and business plan.View the profile
Product manager in AII like creating things and writing about them!View the profile
About the talk
Large scale open source projects can be daunting, and we want TensorFlow to be accessible to many contributors. In this talk, we will outline some great ways to get involved in TensorFlow, explain how its design and development works.
Presented by: Joana Carrasqueira, Nicole Pang
Welcome everybody. It's been absolute pleasure to be here with you today has just calling mention. I'm trying to cut his crate and I'm a program manager for tensorflow it Google. I'm joined by my calling me call Penn Nicole. I'm a product manager for 10 so I could go and you're going to talk about this in Suffolk Community. And the many exciting Ways by which you can get involved in the work that we do. So let me start by saying thank you. Thank you to you. Thank you to the community for all the hard work that you've done this with open source tensorflow in
2015. We first see so many contributions and so much support from the community that really the project where it is today. Is it to you to all your efforts and all your hard work? So thank you for that. Just Encore tensorflow alone. We've received more than 6,000 commits from over 2,000 contributed. This is so impressive but not just only this on stackoverflow. We have received more than 50,000 questions and we have on board is more than a hundred and twenty machine Learning Express 2 hours Google developer expert program
established fuse to use the group's all around the world. We've also has 1525 gas post on our tensorflow block. Which is fantastic and opportunity only continues to grow here is a snapshots where you can see that the number of commit from 4 years ago as being rapidly growing under so much support and excitement from the community. We truly couldn't have gotten this far if it wasn't for you for the country Buddhist for all the words that you do. So, thank you so much for that.
And it's not just the contributions. You see in the feedback. We get from our community on GitHub and stack Overflow. But of course I miss you all know tensorflow is the global global worldwide community and we see a lot of love for ten to flu also on other avenues and you probably have heard a lot about TF2. I do today yesterday and you really will hear more about it tomorrow, but his is one instance. Where are Global Community response really positively and we see so many cases of that and today we'll touch on these cases and of course how you can get involved in our
community. So briefly what we talked about today. We want to tell you how you can learn tensorflow how you can get started on your own journey of using tensorflow, whether you're more in the beginning stages or you're really Advanced user a pencil glowing applications other we want to showcase to you our Global Community. I'm really run through some really amazing use cases really really tell you what we've seen people do tensorflow are people use tensorflow for and hopefully that can be very inspirational for all of us right in the community. And of course what you're here today,
you want to know how to get involved with tensorflow. So we'll walk you through not just the way that you might first think of which might be contributing code because I'm supposed open source, but also a lot of community groups a lot of special interest groups and those again are all over the world so both for a volunteer in this room, and of course everyone watching online and there's many many resources and we're so excited to to share with you. Which really has you could say which really have a vibrant Global community that
continues to grow because there's so much that you can do so much that we can all contribute to tensorflow and let's have a look at where our community is space. And what are what are they doing right now? So it does the flow user groups. It's they are a wonderful way in getting involved with tensorflow either online or face-to-face. You can meet with other like-minded contributors and developers really to answer questions to solve problems challenges. And if he building those cases on really how you can Implement tense of flow across different industry. So
just an example one in Korean and that one is the biggest that we have in the world and we have engaged more than 46,000 members. It is very impressive and then it's kind of loan is the country with most user groups and they have you Routes across 15 different cities. It's really impressive how the community is growing so fast all over the world and one of the key message is that Nicole and I would like you to retain from our presentation today is that if you don't have a user group and wear your base or
in your region, feel free to share to start one share your experience disconnect with all the lights minded developers and start talking about tensorflow. We are here to support you throughout this process and attorney. So feel free to reach out to us very happy to to guide you through the process. Like I mentioned if you'd like to start your user group, here are some of the resources that you that you can have a look online. If you are interested in starting your own group, we are so I sharing our Alias so you can really get to know the team and how you can start
creating your music group. Yeah, so in the spirit of honoring our Global Community, we want to briefly touched on eye with a pencil team has been doing worldwide. So like Joanna just sad right we have so many user groups and you really can see on that. They are Global and if you heard this morning in the keno on the tense mood team was really excited and really lucky to be able to go to many cities. I meet many of these users and meet many of the company than you may need to start up. So using tensorflow at me so many different cities in the world. And
of course we're so excited that you're here today from the temperature in Santa Clara on today, and we're really really excited to again be able to see the use cases and we'd love to share briefly on some that use cases with you. The first half when we look at Asia and Asia Pacific and there is a really big Vibrant Community there and that's when I decided to write a lot of people in Korea a lot of people in India lot of people in China there. All you do with two amazing applications do in China for instance tensorflow is actually not just acted on her applications. But also the community is
really active on our official types of a WeChat Channel and Disney Channel show cases. A lot of use cases of TF light on mobile. Like you can see this one example of a video pot from called it with image segmentation and on mobile devices really awesome work and not just doing awesome work but also sharing right with all of the community on on the WeChat blog and we're really really glad that we're partnering with them and really glad to see these use cases. Yes, I need to call and I were really fortunate that we are able to join the road shows and really connected to
local communities worldwide. So for example at the Roadshow in Latin America, we connected with the letter which is a startup in Chile and they are trying to detect supernovas and galaxies to the users of a child processes and machine learning is really cool and they use conventional and neural networks to classify astronomical objects contained in a stream of about 200,000 images per day. The word that they're doing is so impressive and it's absolutely worth sharing with the rest of the community another example in Europe we
connected with IM which is a library of photos that uses tensorflow for object classification and they're all the Rhythm scores photos based on their static quality. Also on the on the relevance to your Brand's visual identity and then every photo is automatically tags with keywords just to make sure that the entire library is searchable. It's really impressive and then they use at SFO lights on mobile to make it easier and also more responsible for the users to to use and I
am and then lastly in Africa with we met with many exciting startups are trying to find a solutions for problems that a global scale that were relevant to to the region, but we would like to highlight the Great's BoRics off the blue Health full averages the power of machine learning and spectral analysis to really turn any smartphone into a powerful non evasive at screening tool for pneumonia asthma and other pulmonary diseases. Neural networks for modeling
spectrograms that were generated from algae analysis through their smartphone and then the series models and they're frozen and converting to the floor light and the converted model is done deployed to a mobile device to perform interference of the cases that we connected with during the Roadshow and it was brilliant to see old is very innovative ways that the community is using and building and around since the flood. So this were just a few pictures off power Road shows where we truly engaged for the community and it's possible. It's very tangible
the excitement that we see not only from contributors but also from users of ostensive flow. It's fantastic to see how many of the start of another companies are truly impacting and changing the world and this is all you using tensorflow. Thank you for that. Yeah, so I would like I like we said in the beginning, you know, we wanted to do a brief overview of some just just a very small sample of some of the awesome use cases of tensorflow, but I'm really digging to what is available for you. Right? So one of the first two letters that we'll talk about is
education now why is education important for us a tense of flow? And also we hope it support for you and the community well tensorflow is of course, it's you if you are very very very hard to make sure learning resources are available to everyone in the world and we really value. I'm not making just the product better for Learners. So for instance TF2. 2. It's easy to Buggy and the usability of Tara's it's designed for that for that better experience for Learners. So not just the product but Also the educational resources so love to go into more detail this morning.
You heard about our launch of the new learn ml Hub on to throw. Org. This is a great tool because we heard people feedback that they would like more curated resources on his foot at work. They would like more passive learning from know whatever level of machine learning and deep-learning all as you have into more advanced applications of tensorflow. So we heard you and we not responded with this new resource is not just a compilation of curated at resources guy that
passed, you know, whether you're a beginner weather in more advanced than Which Wich Wichita Orioles what guys might be helpful. And also if you are interested in tfjs on the browser, we have a very detailed very very nicely organized. Learning Resource there and we hope that you'll progress through it and whatever stage you are. If you are more advanced with tensorflow, you might still be interested in R&R moves are massive online multiplayer online courses as you probably already know tentacle has great relationship weight Partnerships with both deepen AI for Sarah and I'll see
you. City and these courses are available again to everyone everyone in this room to room watching a lion and we really hope that you'll take stuff that we having these courses which is both from his phone instructors and also renowned academic instructors to and we we really want to give everyone a ample opportunity to to learn tensorflow. And is he heard this morning? There is a specialization on Coursera for 10 slow data and deployment a really taking modeling not just understanding how to build a model but also deploying it in application. And again, right as you move
out the steps of knowing tensorflow. We really hope you'll check out our new and updated tutorials and guys. This is due to the things to the amazing work on our tents. Relations team. They're constantly writing a new documentation you guys new tutorials and where the launch of tattoo. O all of these new guys are available for for you. Check out tattoo. Oh and really understand how to use carrots and really understand all that use cases. There's some really amazing detail documentation here. So we really hope you'll take advantage of these resources that we we provide.
And finally, let's jump into how to get involved. So now, you know tensorflow write your answer as if you want to be contributing to open source Community. Well, one of the first ways that everyone kind of things about is contributing code and we're happy to describe to you away that we we use at 10:14 to cancel widely was supposed to sign into us a TI designs and also some of them are driven by by community members it says request for comments are rfc's so this is actually the main main way we communicate I'm changing. Sorry if he is ever
receive design feedback. So we'd love to invite everyone here to take a look and also showing This is one example of an RFC. This is an improved RFC tens of flow sensor Forest estimator and I would like to take this opportunity. Of course, thank everyone who has authored or reviewed an RFC. And we actually have 45 rfc's accepted today, which is really an incredible incredible number and they have range from TF an HRC expands the usage of tensorflow, right? It really
helps the community and it is also as it is a great Boon to to attend field team. So we'd love to have you also proposed to die in so you can check out more about our season. Of course. I talked to any of us about this also. And and also for bigger projects in which we have to work as a team with created the special interest group the seeds out which is the program that organizes the country Buddhist into more focused dreams off work and everything started to be distinct bills and nowadays. We have 11
6, which is really impressive. How does steak have also and thrown so much over the past few years. So all the country Buddhist you are very welcome to join two steaks and religion just saved as raising a s'more with a parts that you either enjoy or care the most about tensorflow just an overview of our country better echo system as you can see highlighted in them orange how we have this week. I don't think bills I owe networking gvm and micro and run. Which arm Community
lead open source, six and the others which include carrot carrot Swift mlar and since the board they are Google lead with an open design philosophy. So if you see that resonates with the word that you do or if you if you care about the topic and would love to learn more mistakes have monthly or weekly calls and you're very welcome to join us. Well, I would like to give you an overview of our open source Community less sick and just briefly going through some of the key aspects of this
takes the stick add-ons it maintains important additions to tensorflow and ended up with some of the parts off tfcon trip and this steak is led by Sean Morgan and Sue white tongue. the stick built we have one after the lease actually here with us today actually focus on building and package tensorflow for different distribution in 5 minutes and is led by Jason Zimmerman and Austin Anderson a focus on supporting extra file systems and file formats for test and his initiative led by Young tank and then Anton Demetrius I'm unhappy old note high performance
Computing resources. They require lightning-fast interconnectivity. And the thing that's working Focus exactly on that on building more Network supports for tensorflow. This is an initiative led by a bear in and show him the door. I'm finally the speaker has I wish we could just take care is to continue to improve the carousel API for for tensorflow. So those are some of the things that you can join but we also have like I mentioned before the other states that are also Google lad but with an open with an
open philosophy, you're very welcome to have a look at the stick Playbook at the Center for the Arts where you'll find more information on how you can join the steak and an ongoing project at that they have right now if you feed that are none of the states that currently exist are a straight for you or for your work if we see if there's enough evidence and enough support from the community. You can also start and establish your own sake and if you have to get hot on our Community Resources, that's where you'll see how this place operates
one of the resources and tools are available for you to help you throughout this process Federal. So we have more information not only about the safe but also our IRS to process and I work with Contact I strongly encourage you to have a look after a citizen of the world. And I'm also extremely excited to announce that we achieve another Milestone. It's types of flour in our community. We have hosted the first country bit of time it just on Monday and Tuesday for almost a hundred participants and it was a great way to really connect
with a sigelei and with a with a broader community and to really understand how to gather we can move forward with you open source project. What are the dates that the strategy the Strategic development so we can Implement in tensorflow. What I did the communication need a project management Community Management. It was a great conversation. We have over today. So I strongly encourage you to if you didn't have the chance to participate it's time to have a look at you online resources that will be available after It was a great opportunity to connect with you all.
Awesome. So some of the vi likes you and Richard are lied by I will be calling machine learning Google developer expert and so I'd love to show you a little bit about what that means. So are MLG D's are a Global Network of ml experts are the Google works closely with and we provide latest information tonight. So we're really excited. We have a hundred 626 MLG D's to date worldwide and this year alone these MLG D's have given over 400 talks worldwide photography workshops worldwide and also written over
200 articles and this is incredible because we actually know that these talks workshops and articles have reached a worldwide audience of 435 35000 developers. Does he can imagine tensorflow our team we want to reach as many people as we can but with the MLG D's, we really just Amplified. Which in fact that we can have in the world of teaching it at 10:30 flow and really really helping people all around the world entertainment sensor flow. So we're really excited. We would love to tell you if you want to become a TD. This is also a link to become a
GD. We also have a lot of links for I'm connecting with other GDs and today would also love to welcome. One of our GDP is up to the stage to give a brief chat with us, so please welcome Jason Vaughn turn my back needs for some bills. We have a few members of build around here. Thank you. And the bills being the first dig it was from 2 years ago. So I really seen a lot in that time. We have so many cigs we all kinds of things. And they really started like I started build
because I saw problems with and I was trying to use it and I want to make it better and like really the group is grown. and Do a lot of great things. I want to encourage everyone to get involved. You can join us if it already exists. You can find it when you want to do work on it by more people that also just said maybe starting you said a lot of people around to help. These people are wonderful. And I'm also one of their energy he's so great program. It's really nice to hear. I like from other emojis.
They work on all kinds of cutting-edge stuff all kinds of different fields stuff that I don't even know or hear about other than them. So really good. Thank you. Thank you so much a person and another were really lucky to have another mltd is an audience and please welcome Margaret Maynard Reid. Hello everyone. I'm a machine learning GED. I'm also the lead organizer of Google developer group Seattle and another group called to Seattle data analytics and machine learning. I became a missionary in GTE in 2018. And here's why why I
love being part of this amazing Community. I get to collaborate with other Machinery in g-d's and Google Earth on various projects. For example, I get to write some tutorials and you will find out has a flow. Org and some of the blog posts that were published and they tend to flow medium publication and earlier this year. I helped organizing the global test a spring with page Sergey and other Machinery in g-d's and Gigi organizers. It was an incredible experience to work on such a high impact project that which was Even mentioned in the
keynote this morning. So I speak about tons of fluid on device machine learning at various conferences. And I really enjoyed opportunity to be able to preview Google products and provide feedback. So many of the Machinery in g-d's are well-known Educators speakers or O'Reilly book authors. It's really great to be able to learn from my fellow GDs and Google Earth and why is he here? We'll gather together for our Global GDP Summit around the world that he's from around the world and we just had to submit a few days ago before
tensorflow world. So to become a gde Machinery in GD in particular, you need to be able to demonstrate both your community contribution as well as knowledge e e Mission Ernie, we would love to see more of you joining our growing machine learning gdh mean Thank you. Thank you so much Margaret. This is fantastic. I am I am sure I can speak book for both of us. But I'm always so impressed by the fantastic and amazing world that I would see these two. It's really nice to see how it is
engaged to communities all the ways by which you can contribute to tensorflow. It doesn't have to be only true, So if you if you are a colder, but you would like to learn or develop a new skills that there's many other way so you can get involved with with tensorflow it when we when it comes down to known cause contributions. There are three main pillars that we normally encourage our contributors to join at primarily on used to support which includes creating documentation translation training
courses that really will help other contributors. I getting involved and on boarded within the project. Community management related to organizing it sends me top and old initiatives that's get the community together and energized and excited about machine learning and tensorflow. And then on the project management, I'm creating the tools and resources that will help Advance our project but also keep the house and the sustainability of the initiatives that we do sometimes you work really on cross functional team really building to use cases on on how
tense of flow can be implemented in different ways. And then finally I would like to highlight that we have a code of conduct in our tensorflow community. So we applied is Code of Conduct to all the events and the initiatives that we do and we would like to remind you that this is a safe space where you can truly be yourself as a contributor and very welcome. Diversity off. Opinion and suggestion. So if you see that something is just not right to please feel free that
you know that you cannot collect those problems too old to the communities to be here for you. We are here to make sure that you feel engaged that you feel hurt and that you feel that you belong to a community of excited machine learning experts contributors and Jesus. So we want to wrap up our conversation by revisiting the lynx and a different resources that we've giving you in this talk. So again after two worlds, you know, you're wondering how do we keep up with the latest news in the latest? I think the hives from tensorflow what
these other ways you can keep up with us. So of course Twitter is very great for a lot of the latest announcements and updates from the touch pro team. Our blog is actually an amazing resource of a lot of deep dies a lot of like understanding these specific use cases. You might be wondering how to use the following is an application and the blog may actually have a guest post or post from the Consortium that can dress. So are we really really stress you check out the blog? I think many of you probably already seen the YouTube channel, but in case you haven't it's actually a really awesome
resource to learn tensorflow. So we have a lot of videos that high. Light again, like our new announcements Riley how to use tensorflow how to use a specific things like t s Cara's we have videos about that and one of our most popular videos is actually down by on someone are Developer. I really should seem Lauren and it's the MLS Zero to Hero video and it's a great resource. So again, like we haven't seen the social resources. We really highly suggest you follow and and that's how you get updates from tensorflow outside of calf world. And finally, I'm just on the links that we
showed earlier. I we really want to emphasize again tensorflow. The community would not be possible without everyone in the room near without everyone in the community globally. So we really encourage you if you aren't in a sick or in a user group, if you're interested, you can check out everything on art as well. Org community links are you can check out the educational resources. I mentioned at the beginning and we are so excited that so many of you are in the group today. So we'd really love to you know, welcome you to also share with your fellow a conference attendees. What is like being
in a cig what is like leading leading us a verb an MLG D2? So with that we have our call to action to you, which is John Dee's Ruth join the 6:50 part of the community country with code 210 safflower documentation translation and occasional resources event. There's so many different and exciting ways to contribute to sensor flow. So thank you for being with us today. It's been a really really a pleasure speaking to you about the many ways. They can get involved with the community and I hope that we can
Buy this talk
Access to all the recordings of the event
Buy this video
With ConferenceCast.tv, you get access to our library of the world's best conference talks.