 Hello, hello, hello. My name is Marky Jackson. I am one of the mentors for the Jenkins Google Summer of Code machine learning project. I want to welcome everybody for being here. It is Sunday, March 22nd. The time, Pacific time for me is 5 p.m. I would like to let everybody know that this is a recorded call. We are under the Jenkins Code of Conduct, which amounts to essentially just be nice to everybody. We have a very large group here. I am very very excited to have everybody. I'd like to inside the Zoom chat, I have placed a link to the document. And if I could just have everybody go in there and add their name under the attendees section, that would be great. I will be taking notes for this meeting, but in future meetings, myself and the other mentors will be kind of bouncing back between me note taking and what have you. Marky, I want to mention that maybe other people cannot add their names if they don't have any rights on this document. Yeah, so I have just also to get into that. I've made this document a comment only document as we move into the student selection process. This document will be, it will host all of our meeting notes and that will be open to everybody. But for the initial part of this, I decided to make this only comment only except for the mentors. Mentors have full ability. So if you see that there's a comment in there and you approve, please feel free as a mentor. You know what I'm saying and Bruno to go ahead and just accept the comments as you see fit. Awesome, awesome. So I see everybody's put it in there. I want to welcome everybody to the machine learning project. I kind of want to give a overview from my, I'm going to put my org admin hat on and sort of given a brief overview of what the Jenkins project is about and what this project is about. And then I'm going to turn it over to the mentors to introduce themselves. And then I'm going to turn it over to the students, the proposed perspective students to introduce themselves. And then Bruno and I'd like you guys to give an overview of the project after that and then I'll come back to explain the timeline. The proposal submission process and then as mentors what we're sort of expecting and then we'll open it up to the for sort of a Q&A. If everybody is fine with that, I'd like to say, first of all, welcome everybody. This is the 16th year of Google Summer of Code. This is the, I'm probably going to mess this up, but I want to say it's the ninth year that are six year maybe I'll get back to you on that number that Jenkins has participated as an organization for the Google Summer of Code. The machine learning project is a project that we tried to get off with the ground last year and it had a lot of good success and I think there was a lot of interest in it. Machine learning and data science is one of those fields that has gotten a lot of interest and I feel that personally that this is great for students. With that, I'm going to turn it over to the mentors to introduce themselves. Again, my name is Marky Jackson. I'm not only a mentor, I am also one of the Jenkins organises. This is my third year doing this. And without further ado, mentors, take it away. Okay. This is Yanis Moutachos. I'm in Boston, Massachusetts. I have been working with Jenkins for about 10 years now and always have been using it in the context of scientific applications and data processing and data analysis. And this year I'm again excited to have the project on board with my colleague Bruno from New Zealand. We've been trying to advocate for Jenkins as a data analytics platform. And I'm really excited to have a lot of good interest and good proposals that we're looking forward to to entertain this year. So maybe Bruno can say a couple of things. Sure. Can you guys hear me? We can hear you loud and clear Bruno and it's great to see your face. I just brought my computer from the office. We are shutting down today in New Zealand, too. So I'm Bruno Kinoshita, originally from Brazil, now based in Auckland. As you already said, we've been working together for a few years now, I think probably more than two, more than three. Trying to bring more people to use Jenkins for not only DevOps and continuous integration, but for other things as well, because it's a very good general purpose workflow engine and can be used in other areas, we think at least. And we've been using that with some success. We have a few plugins that we maintain, both in the Jenkins organization and outside. So I'm really looking forward to what these students will be creating with Jenkins. And I mean the time zone of plus 12 or 13. So that's why sometimes I miss the these meetings because it's my work time. So I just take a late break as long as there are no meetings. But feel free to ping me at any time on GitHub or GitHub and I just may take a little while to reply. Thank you. Thank you very much. Let's go ahead and get into the project overview. If you if the other two mentors who would like to give a project overview, I've linked in the notes, a link to the project details. But I'd like if you guys want to spend a few minutes talking about what the project is about. I think your analysis is probably the best to explain because he has a he has actually a use case where he can apply some of that. Sure. I'll be happy to I'll momentarily turn on my my video here so the students can see me as well. And I hope we don't create any issues with the with a zoom meeting, but the idea is that we created tighter integration between Jenkins and and some of the machine learning tools that are very popular these days. Typically, these are the interactive notebooks. And as we know they have a back end that is driving the calculation of each step in the notebook. And it's really not that much different from what perhaps build steps and the job parameters in Jenkins do. We have examples that we have developed using a plugin called the active choices plugin that uses groovy as the scripting language. And in this particular case, we are driving external calculation agents such as the our language and even Python various image processing tools such as image magic. And we're able to create quite an interactive user interface for Jenkins before even the build starts. And this is a kind of different from what Jenkins usually is designed for. Jenkins typically has very little upfront interaction with the user interface, although we do have freestyle jobs as they are called where you provide parameters. But this is a very similar situation as to how you would provide an adjust parameters in in a notebook. And then you would be able to do calculations during the build step and so given that interest and given the fact that we want to introduce two more data scientists Jenkins as the platform and workflow engine for data analytics. We thought this year would be great if we can provide a more formalized way for essentially linking the back end calculations which are the kernels that are first of all they are multi lingual as we know they can work with Python they can work with are they can work with Julia they can work with groovy for that matter Java and a number of other languages to perform machine learning calculation. So the idea would be to try to integrate these computational kernels that typical interactive notebooks use with the Jenkins build process and this way we have a tight integration between the two. Anything else you want to add Bruno. No, I was exactly waiting to see if you are going to point this pain point that we have with the active choices there is is very easy to run a workflow with with multiple language. I mean with groovy and maybe with Python are but it's not very easy if you have an existing pipeline that our researcher created for example in his notebook. You then have to translate that to Jenkins and try to match what you have and use the existing plugins for that so that's a huge improvement if we can use either similar tools as the researchers are using or connect via Jenkins and maybe having having just one more step there is to call from a built the notebook. So it's exactly what you already said and hopefully it will be something that the students can help us achieve. Yeah, that's a great point. Thank you very much for that I appreciate that. Moving into the next step I wanted to talk about the Google Summer of Code timeline. As you know that the 16th of March is when we started accepting student proposals that will go through the 31st of March. And then up until April 14 the student swap request will be due from the org admins and that's where we will actually review. So we are actively in the review process. So what does the review process mean that means that you should already kind of know what the project idea is about and you starting to work your proposal for that. I want to say that we submitting your draft proposal is done to the Google website. If you are in the getter channel I've made a few mentions about that link and where that gets submitted to and at that point once you submit your draft proposal. You let the add or let the mentors know that you've submitted it we will start reviewing it and maybe request changes or say it's great and then you will change your proposal into the actual full submission. Does anybody have any questions about the proposal or the proposal submission process. Mike it's Bruno. So the question is regarding our job as reviewers. Do we have any deadline when we should submit the review back to the students. I think as as mentors and reviewers we should be reviewing these sort of live. So as a student tells us that they've made a submission we should go through and so first of all let me take a step back. What students should be doing is your draft should be submitted to Google but you should also have that in a Google doc in the getter channel. In the getter channel I've also said that make that reviewable with comments or student make that available via comments so we can comment on your proposal. That process takes place between now and the 31st so I know we have a few proposals in queue and the mentors we are reviewing those and we will make comments back within the next I would say three to four days. Do any of the potential students have questions regarding that. I am typing a question that I think it's relevance. And so everybody can see it. If you can address that point. Yes so your proposal in draft form is not available to anybody else but the mentors of the project. That's why we ask that you do not only the Google draft submission as well as a Google doc so we can make comments on that doc. Once you make your final submission the final submission is public. Which also probably means that they should not be posting links to share documents unless they have the sharing sort of configured so that only the mentors can can review it at that point. Students have the students have the they have the ability to share the doc with us directly. And you don't if you do not want other students to see your draft that I would not share that in the channel. I would say for trying to figure out that I want to do this in a way where everybody is being transparent. I understand that the under the sort of mindset that students you know you don't want other students to know what comments are being made and then they can change that. If you do not want to share your draft publicly send it to one of the mentors the mentors are all in communication with one another. So if you share it with myself I'm going to let the other two mentors know this is the document and then I will reach out to the students say I need you to share it with these other two. The other option is is to if you do the document only to the mentors is to reach out and say hey mentor what is your email address so I can share this with you other mentors. Yes, I think I would I would encourage that I believe our our email addresses are all on the document that we have shared the proposal and so people can add the three of us as reviewers to their documents and share it with us. Please do make sure that the the when you allow the sharing that it's only comments not editing rights. Is there any other questions about proposals. Or the submission process for proposals. Hi, everyone. Hello. Hello. Welcome. Thank you. I have question about the proposal. So I'm having some concerns because I've been trying to work on the proposal for some time now and I've been wondering like if like I'm not really familiar with a Jenkins so since I got interested in the project I've been trying to learn how it works. So I understand what we are we have to do in the project but the thing is how to write detailed implement implementation inside my proposal and it seems like it's taking me so much time to understand how Jenkins and the plugins work in order to able to write the proposal. So the question I'm asking is taking so much time like that. Will it be really affect the time you take me to write the proposal like it seems like I see why don't really understand in details of what I'm supposed to do in the proposal. So the proposal is really supposed to be about your logic in implement implementing the actual plugin right so it's not so much about writing the code. If you do that that's great but most often it's going to be about your logic and the understanding of how to implement the project so for example. If you've never worked with a Jenkins plugin. The first part of your logic and then I'm just sort of giving feedback to everybody here. The first part of your logic is going to be how to understand to implement the plugin. How to work with the special Java classes and staplers and things like that in your given ID to implement the plugin. The second part of your logic is OK now I've got my little you know my hello world plugin working. Now I want to start to extend that to a Jupyter notebook right so you have to understand that's what I need to do in your proposal you have to write the logic of how you're going to do that. So I gave two examples of one I need to understand the basics of implementing a plugin so my first week and I'm just sort of saying this ad hoc my first week. I am going to in my schedule part of the proposal. I am going to understand every piece that I need to know about the plugin and maybe day one through three. I'm understanding how the stapler. A portion works for plugins and then you start to work from there but you're building out how you're going to do that doesn't mean you're writing the code to do that. You have to explain your logic for what you're going to do. Did that make sense. Yes it does. I would like to also add that perhaps you know because we know that this is also kind of a learning process for students to engage with Jenkins perhaps for the first time. So it's not necessarily unusual that you don't have any experience yet with Jenkins and the way the plugins work and so on. But as part of it. The initial proposal I think will be at least you know reading some of the reference material that exists and for that matter actually Jenkins has very good documentation and understand a little bit the nature of plugins. And at least the tools and the language that you're going to use for that. And you know the mentors I think should be able to also do we be sort of this review of the of the document perhaps give additional guidance in terms of the types of information that you will need to be successful in the project. Agreed. Like I've noticed that there has been some questions about setting up sort of your plug in environment using sort of that hello world. And I ask you those questions is OK. And again I'll go back to the what I was saying earlier. Writing your proposal is not so much about writing the code. It's explaining what your logic is to achieve your schedule to writing code. Do any other potential students have questions regarding Proposals project submission or a proposal submission. I have a question that maybe have been discussed at some point but I'm not aware of it. Does Google have any Plans on extending some of these timelines for the submission or are they still sticking with the end of the month given the situation with the coronavirus around the world. At the beginning of this week. Google did release a statement as far as Google summer code and the students submission time frame and it was said that they would not extend that submission process. So we are still abiding by the March. That's good to know. Thank you. I have another question. Yeah. At the end of the submission of the proposal and acceptance. How many students we accepted for this particular project. Can you repeat the question. Will they be accepted for our particular mission learning project. I believe it's only one. That is correct. Yeah. So between the third the end of proposal submission is 31st of March. April 1st through I believe April 14th is when we will. We will notify who which student has been approved for that. Is there any other questions regarding submission proposals. Okay, I'm going to move on to the next section of our notes and that is what a mentor is. I. I take this part I'm going to start with this and then I'm going to turn it over to the other two mentors. And the reason that I wanted to start with this. I'm going to try to do this the right way. I'm going to just say it and I hope I apologize if this comes off the wrong way. I spent this past yesterday actually talking on a call with the Google devs in Africa. And one of the things based off of the messages that I received personally. One of the things that a mentor is is we're here to guide you. We're here to help you. We are here to do what it takes to help you succeed. And on that same token, we are, we expect a certain level of research before you're asking a question. So for example, and I'm going to just use a random example. If you run into a problem. My expectation as a mentor is you've exhausted your personal efforts of research. For your given problem before you've asked a mentor for help. A good example where did I, you know, I have a stack trace error. Did you, did you Google what that stack trace error was. I have, I need some questions I've received or where, where can I find the project documentation. That's a fair, that's a fair question. One time. The fifth time that's not a fair question. I want to make sure that potential students understand that mentors. We're doing this as a volunteer efforts. Most of us and I can speak for myself and I think I speak for the other mentors have day jobs. We expect students to do their research. I want you to feel that, and I'm only going to speak for myself in this part. I don't want a student to feel that you cannot come to me with the question. And there's no such thing as a bad question, because I can guarantee you I've asked it. But at the same time, I'm expecting students to have done their due diligence. If I, if I get asked a question that I can Google that same question and find an answer. I don't think the student has done their due diligence on the same token. I want you to feel comfortable enough to ask, but if that becomes a repetitive thing. What, you know, as we do the reviews through the, you know, the, the coding phases, we will start to look and that becomes where I can say from past Google summer codes as a mentor on projects, where we'll start to evaluate that the student is not doing enough of the due diligence needed. I'll turn that over to the other mentors if they'd like to add to that. I mean, personally, I think what what you said is very reasonable. You know, we all have, you know, busy, you know, work schedules and this is a, a, you know, volunteering position that we want to. Nonetheless, we want to do it and, and be of value to, to you in terms of learning about Jenkins and learning about open source. More specifically, it is also my expectation that, you know, at least a reasonable level of free shirts gets done on your own time to figure out problems and approaches to these. On the, on the other hand, I think we're three mentors, which is a lot rather large number for these projects. So you should be expecting a reasonable interaction in time with us, maybe not necessarily with each one of us, always the same person but with us. And all of us are pretty excited about this project. I think you should expect a reasonable level of feedback and, and, and assistance, reasonable assistance. Yeah, I agree with what you and is and mark said and it's similar in a company when you are a junior engineer and you join a team. But coworkers will have the same expectation they will try to help you as much as they can, but they are not supposed to do the work for you. They will be happy to receive any questions that you have. But as Mark said is a good example. There are cases where the junior engineer does not even search for the stock trace. Of course, there is also a learning step that someone has to explain to him how to look for that, but we expected to be proactive, but at the same time, not to be scared to ask any questions, which I know sounds very ambiguous but I believe it works. Just be excited and try to communicate with us and we try the best to help. We cover, I think, most of the time zones from the US to New Zealand so Asia, Africa, any continent can can ask and we should be able to reply within one day, I think. I agree and thank you both for that. The idea here, at least from my perspective is not to be offsetting. I want every potential student and the student that gets chosen for this project to feel comfortable to fail, because you will fail. The idea here is for us to help you, the student, understand how to not only be a better software engineer, but to understand, as Bruno was saying, we cover these three different time zones. We also have a vast, between us three potential mentors, have a vast wealth of knowledge. We cover academia, we cover startups, we cover enterprise. I mean, we cover a lot. So you're going to get a really, in my opinion, awesome breath of knowledge. But at the same time, we expect you to to be excited about this. We expect, like what I look for in a student is I look for gusto. What is gusto? Are you excited? Are you going to go the distance to do something? When you come to me, it's going to be a problem where we can collaborate and I can help you and you may be even helping me on something. That's where great things are built. But if a student comes to me and I can Google a question that they're coming to me with, and I can find the answer, and it fixes the exact problem that I'm able to duplicate, then I don't feel that student has given. I may not cover everything. So please don't think that that's going to be everything. And this is not meant to be off-putting to students. But I just want to make sure that we have an understanding of what the mentors are here to do. I also just want to clarify that what perhaps you were saying earlier, we don't expect you to fail in terms of the project. Yeah, I think it's more in terms of having challenges along the way. Thank you. We want you to succeed on the project and you will. Agreed and I apologize and I'd like to make an amendment to that statement. I don't want you to. I personally do not want you to fail at this project, but I want you to as a student understand where you may come with something that may not be the right way. And as a mentor, we help logic pivot to the right way. And that's really what software development and the larger picture is. Helping you to understand what it is like the logic of how you're going to do something you may want to write code is certainly but there may be an enterprise mindset that helps you. If you have all of these visions, you may realize that there's a different way to do things. This is to prepare you as a student for the bigger picture. If that makes any sense. Does anybody have any questions about what is the expectation of mentors or I said a lot and I know there may be I hopefully I don't seem so off putting that's not who I am. I think most people that know me know I'm not that type of person I'm very empathetic. So if there are any questions, please do do let me know that they let the mentors know if you don't feel comfortable reaching out to a certain mentor reach out to the to the other mentors and even then if you don't feel comfortable reach out to the org admins. With that, I'm going to move into the open Q&A. So this is a this is a time where and we're going to do this every week. So this is a time for the students to just ask questions. There may be lots of questions I think in the beginning, there may be lots of questions I think as we move through and student student is chosen. There may be less questions, but this is the time please ask questions if you have any, if you don't feel comfortable. That's totally fine as well. You're more than welcome to reach out to one of the mentors. Yeah, so I just had a question. So my question is basically more related to the project as in from a specification point of view like the exact user requirement. I did go through the project idea. Basically what I what I can't envision is the final fight the final plugin or the final product right I mean, because when I think about Jupyter notebooks, when I'm using a Jupyter notebook what I'm basically doing is I have all these cells right and I'm running my Python code and I'm going above and below just running different cells and running my deep learning or whatever models I have it when I'm running Jenkins. My idea is that basically maybe you can think of it like like how I think of it is that you know I have some tests right and I need to be able to continuously deploy my application. So like I don't understand how this product matches that product like what exactly is needed by this project. I'm sorry if the question is too naive. No, no, no, it's not and I actually think that as as mentors and Bruno illness correct me if I'm wrong we've had discussions about which way this plugin should work. Yes, Jenkins control the the sort of push to the the the Jupyter notebook and that underlying kernel, or is it the opposite way. We've had sort of. So if you think about like command and control, where is the control plane. Is it initiated from Jenkins, or is it initiated from, you know, a Jupyter notebook, we've had this discussion. And that's a very good question because I think I saw a document or something on the get get our channel. And I was surprised when I saw that that it was displaying the notebook within Jenkins and that's that something that can happen to the use case that we have. It's more with customizing the runoff a model in Jenkins. So the control is with Jenkins. What we we had in mind in the proposal in the final proposal after we discussed and you and is a market and correct me if I'm wrong was more in the sense that, instead of executing the groovy or our code in Jenkins, we would execute that in a kernel, which allows us to do something similar to the scripted script killer plugin. But instead of having to install the plugin and go through security, etc. We would have the ability to run the code with a Python kernel or Zeppelin interpreter. It doesn't mean that we cannot go forward and do more stuff like executing a complete notebook, embedding the notebook as the HTML publisher plugin does doing something similar and storing the output of the notebook in Jenkins. And for running the models that will that will at least I think you and it has a good case for that. I don't think we need to execute a notebook all the cells step by step and get the notebook output HTML input in Jenkins we need more the ability to execute a piece of code that we give in Jenkins, externally, get the output of that code and bring back to Jenkins and then either use the parameter or to put as a build step output. In this way, were you doing something. I remember early in our discussion we were talking about and correct me if I'm wrong, we talked about image sequence sequencing and the image sequencing happened outside of Jenkins but fed the image sequencing back to Jenkins. That's correct. But let me also kind of reframe this a little bit because we put forward a proposal and some of the students I think they have some some good ideas and we don't necessarily want to force the entire sort of preliminary idea we have about the project upon them. If you think that there are some other viable ideas that we could do with Jenkins right now I think that that it would be great for us to to know and understand. For example, I think what I'm talking perhaps about the same proposal that I saw that Bruno was talking about earlier. The idea of having a notebook. Sent to Jenkins and pieces of code, for example, extracted from that because somebody was saying something about a parser for extracting the code. It remind me a lot of kind of the things that I'm doing now. But essentially, you know, what you, I think that proposal was talking about is externalizing the code that's typically embedded into a notebook and it's very difficult to work with because you cannot put it into an SCM repository you can't quite reuse it easily across projects and someone have it perhaps, you know, create these these pieces of code in different languages that are available as built steps or as parameter steps. And because Jenkins does not have the built in capability of doing those computations. That's where we're bringing in, you know, the kernels now and maybe now the kernels can be, you know, plugins that provide this functionality. Or that they can, you know, fire up and create the correct context for Jenkins to execute those calculations. So, but, but it, you know, at a very general level, what we're envisioning is the capability of code in Jenkins in a multitude of languages. You know, from Python to perhaps Julia to groovy or are being executed in an external or internal kernel. And they result in fed back into Jenkins for continuation of the computation, or for presentation of a graph, or, you know, passing that the data set to another computation. I will say one of the things that we were using at MIT was the use of Jenkins and histograms, especially as it pertained to cancer, mainly to breast cancer, and using the histograms to run a model against it but Jenkins ran that model, however the model was in Jupiter, it was a new notebook in Jupiter. So just to give potential students an idea of the different ways that you can do this. And to add what the, to add what the other mentors were talking about is we present the project in a logical form for what we think could be. That does not mean to say that a student cannot come along and say, I see what you're saying but there's a different way to do this. And here's my logic for doing this. If that makes any sense. I think it doesn't. And also the, the, there is a common thing that we have to be implemented regardless of what is the final project, which is this communication between Jenkins and the kernels are the interpreters of Zeppelin. And these only these integration could become a plugin. There is then reused by other plugins and that's something that as mentors we can help the student to do that to package that correctly and deploy. And that could open the, it would open possibility to create a plethora of other plugins on Jenkins, which is really exciting. And so that's why if the student completes only that step for me that will be a ready, a huge success. But I hope the students also excited to see what he's, or she is about to open up in Jenkins and all the possibilities from a machine learning and like from a data pipeline perspective. I see so many applications that like I sometimes lose sleep to be honest with you because when if done correctly. We could even given think of our current world climate with COVID-19. If we were to take sequencing, we could use Jenkins to do sequencing if we had enough data to work with it. But can you think about all those application that to me is mind boggling. Yeah, I believe that the question that Sumit asked earlier has to do a little bit with the typical domain where where Jenkins is used where you know it's Jenkins is just used to deal with code, you know do tests and other things. What we're saying is that we're we're breaking that that that mold. And in fact we have broken it if you can if you can go to the bio Uno site where we have actually, I think we referencing that into the project. There's publications in the scientific literature. There are examples of how we're envisioning Jenkins being used as a data processing and computational platform, and none of that really has to do with the native use of Jenkins as a tool for integrating with source code repositories. With sort of testing frameworks for code and all of that. It's just, you have to step back and simply look at Jenkins as a workflow as a hugely performant workflow engine that can do and integrate anything. And it does not have to be a workflow in the software development domain. Yeah, this project to be honest with you takes many forms. And again, it's like I couldn't agree more it does not have to be purely some machine. I mean you could do so much with this if you if a student really is is recognizing what we're talking about your applications have such far reaching implications. So we are 10 minutes away from the top of the hour. I scheduled this meeting for one hour this meeting will happen weekly at the same time. I do want to open it up if any other students have any other questions this. This is the time to to do jump on. I will also say and I'll spend it just a very small time saying this is when we choose a student for this. That does not mean that the students the potential students that are on here cannot come back and listen. This is this is still open this is a very transparent and very open project and I think the, the greater good of this project is amazing so please just if there's a student on here that doesn't make this. Don't feel your it's because you didn't, you did something wrong or something. One of the things I can say for the from a mentor perspective as well as an org admin perspective is students that did not make it. You will be provided feedback as to why and the hope is is that for next year. You'll try again. And also, the fact that, you know, as you said, there is still transparency that code is going to be developed is going to be in a GitHub repository. The issues are going to be logged into JIRA. The discussion that we're having tonight is going to be presumably open going forward so people if if they feel that they want to learn and if they want to contribute in the future and understand how Jenkins can contribute to data science and data analytics. It's this is just the beginning and you can, you know, the whole concept of open source contributions and development is here for you to to explore. Right. What the students that will participate in the GSOC will contribute is becoming the basis for somebody else to improve upon and fix bugs and so on. Right. For the students that are the potential students that are on here. I can't underscore enough the potential for what is being done here. I cannot also underscore enough the mentors that are on here and their wealth of knowledge, not only from enterprise perspective, but from an academia perspective. This is this is one of the projects I truly am extremely excited about because the application is so wide reaching. Yeah, just as a very quick anecdote when a few years ago we discussed these applications of Jenkins with Koshiki, which many of you perhaps know that he's sort of the original creator of Jenkins. He was, he made a comment, essentially that we're breaking new ground in a totally new application area for Jenkins and he recognized what we're trying to do to recognize how different it was and he was in fact quite quite excited about it. So, essentially that's that's the thing we're breaking your ground in a in a domain that is really hot right now, which is machine learning and data science. We are about five minutes out. Does anybody have any questions. Before you ask your questions if you do I'd like to say to the mentors on this call. Thank you. I understand the gravity of what we're doing, because I think there is such far reaching implications in all forms of academia for sure, but especially in medical science and medical science. I'd like to thank the students, the potential students that are on here that you this is a I thank you for being here. This is a time you're spending with us. I know somebody else wanted to say something so I'll heed back. Hi, I have one question. I've written my proposed draft proposal for this for six pages. Can I extend that to some to add some other ideas on that. I would definitely extend that your document we are reviewing tomorrow. I'll let you know that we are reviewing your document tomorrow. So if you want to add something between now and then please definitely do. And I guess this documents, since they are not do until the 31st they can still be updated to that point. Correct. Agreed. I'd like to also add, please do not submit a final document, please do not submit a final proposal to Google. Until you've had a chance to talk with the mentors, because if you submit the final we can that is you cannot correct the final drafts are able to be corrected finals are not. So with that there. Is there any other questions. Any other mentors would you like to add a closing comment. I would like to thank you to market you you thank everybody but thank you for organizing these it's been great. Market has really done a stellar job and I know this is not the only one that you're mentoring. And so I appreciate your time and help in everything here here. Thank you very much I will say from an academia standpoint this is the one that kind of is nearing dear to my heart. So I thank you, the mentors for challenging me and, and keeping me excited. Thank you very much. With that, if there's no other questions I will go ahead and close this call give everybody back three minutes of their life. We'll look forward to seeing everybody next week. These will be regularly scheduled calls. They are on the Jenkins calendar. These meetings will be recorded and made publicly. We are available at the Gitter Google summer of code machine learning project channel. Please do feel free to ping us with questions. I do not want anyone to feel that they cannot ping us with questions. With that, I will thank everybody for their time. I'll give you back those couple minutes and have a great rest of the weekend and for Bruno have a great week. And thank you everybody. Thank you. Thank you. Good evening or a good day, wherever you are. Thank you. Bye bye. Cheers all be safe. Thank you.