 Again, if you have a lightning talk, please stay there. We are left already. All the people that I was trying to sign last night after the social event, to be first in the list, I drove that to a garbage. And I put a new one this morning at eight. You ready? Yes. Nice. OK. So wait, wait, wait, wait, wait, wait, wait, wait. Go. OK, welcome, everybody. Today I'm Volodymyr Piskun. And I'm going to talk about professional vulnerability of software engineers. Yeah, good. Controls work. So I want to start with a question. What is the ability every developer has? To answer it, let's take a look into what is developer learning loop. When you implement something, you begin with learning new things. Then you implement and receive knowledge to the code base. And then you're very likely to have some obstacles. It is not pretty generoto. It depends on your experience and the complexity of the issue. But we all know it. And you repeat it over and over until we die. No, until we solve it. And there is something which is always there when we are struggling with the problems. And it is frustration. Well, so in a problem, we really frustrate a lot. And while you are growing as a developer, you develop your skills of dealing with frustration. And while we grow, this frustration, which is big and takes our view, becomes something like this. It's not anything in background. When you just so focused that you don't really notice that you're frustrated. And with this, you're really getting better at ignoring your own emotions, which is leading to different problems as procrastination, or anxiety, or even sleeping issues. That could be something else. And one small thing I want to share with you during this lightning talk is that you should never forget to listen to yourself. And for this, you could use different techniques, like journaling, or you can just stop for a second and ask yourself, how do you feel from one to five? And how do you feel now, guys? Thank you. Enjoy the weekend. It was pleasure. Yes. OK. Second one, Marika, Masha, and Lisa from Ukraine. And I'm going to share some. Hello, everyone. We are a small team of junior Python developers from Ukraine. And we are very glad to be here. We have just started our career in software development. My name is Lisa. This is Masha and Marika. To start with, we want to thank all countries that support Ukraine, especially Ireland that opened the country for 40,000 Ukrainian refugees, continue to help Ukraine in humanitarian aspects using every opportunity. This is incredible. And we are really grateful for that. Thank you. How can I? Sorry. I want to share with you my own story about how the war came in my life. On this slide, on the left side, you can see my neighbor's house that I see every day from window in European city. On the right side, you can see something. I don't know how it's naming. That was founded in the yard of my parents' house near Hostelmal. And unfortunately, every Ukrainian now have similar story about his parents, his friends, or his own story. For today, there's no single Ukrainian that was not affected by the war. It's pretty. We want to ask you to help Ukraine. Here you can see volunteer organizations that we support and we ask you to help. The first organization is based on the short item. We are engaged in evacuation of people from dangerous places and humanitarian aid delivery to hospitals, towns, and cities that are in rink or have just been liberated by armed forces of Ukraine that helps with most urgent needs, such as reconstruction of houses of people who suffered. The second is shelter for animals that takes abundant or endured animals from the arrears affected by the warfare. Financially, there is no small help. Any sum is appreciated. And what else you can do for us? Please don't fear to hire Ukrainians, especially young developers, like us. We are really happy to have a job. Most of us started our career at the beginning of the war. But many people in Ukraine are not as lucky. It's too hard to find a job in Ukraine at the moment. So please donate, help Ukrainians and help junior developers, especially women. We are really needed. Thank you. We are open for any questions. So you can find us somewhere here in the hall after talks. Thank you. Thank you. Thank you very much. OK, Sebastian is the next one. What's the name of your talk? Fuck it. I'm not going to repeat it. The red one. The red one? I don't have a little dongle word. Yeah, but put it here. Oh, OK. Give me one second. Yeah, the projector is not happy with the name of your talk. I felt this one was better. OK, we're easy with you, bro. Yes, go. All right, hello, guys. So I'm going to teach you, oh, one second. I'm going to teach you how to fix production code in five minutes. So first, let's start a little bit with who I am. I'm a Franco-Mexican Python developer. I mean, I develop in Python. I don't develop Python. I'm luckily, I'm a numberly data engineer. I'm a Pew Pew lover, obviously. And I've been attending your Python since 2019. So a little disclaimer here. If you use this at your job, you might not have a job for a very long time. OK. All right, this is what it is. So don't you hate it when you push some code to production? You don't test it, because why would you ever do that? And there are some errors that come up, and the code doesn't work. Today, I'm going to teach you how to fix all that. I'm introducing fuck it. Fuck it is an official Python library. You can just pip install fuck it. And then whenever you have a function that fails, you just use the fuck it decorator, and it doesn't fail anymore. Isn't that great? This is good. This is good. Oh, wow. OK. So this package, obviously, is not developed by me. It's developed by AJ Alt. So you should clap for him, not for me. Please go ahead. And as was written in the official GitHub, but you can't see because it didn't load, this module is like violence. If it doesn't work, you're just not applying enough of it. So you can just chain fuck it calls and make stuff that still breaks not break anymore. Let's see some examples. So if you have a bunch of code, and you don't know which part of it is failing, but some of it is failing, you can just use the fuck it context manager, and all of your failing code will work. And if you have a module, and the module is broken, you can just import the module with fuck it, and the module no longer breaks. So yeah. Next time your PM, your PO, your CTO, or anyone comes to you and says, production is fucked, fuck it. So before I knew what this talk was going to be about, I asked my CTO if he would approve the talk. He said, OK, please don't fire me. Thank you, guys. So the PyCon Portugal, so come here. Oh, you have slides. OK, no, wait. Next. Sorry. If anyone wants to announce a conference, there is one minute thing that we can do in the middle. We just like to make your life harder. Yes, after this. Ready? Oh, OK. OK. Five minutes. OK. Hi, I'm Takahashi. Today, my talk title is Spread the Community of the COVID-19 in Japan. Before main topic, did you enjoy EuroPython? Yeah, I know. I really enjoyed EuroPython with good people and nice beer, and nice beer, and nice beer. I'm Takanori Suzuki. Please call me Takanori, because Suzuki is a famous family name in Japan. I'm from Tokyo, Japan. Tokyo is far from Europe. I have a question. Do you know Japan? Wide-hand, almost. Have you been in Japan? Oh, 30%? Do you know Pycon Japan? Oh, 10% to 20%. How do you attend Pycon Japan? One, two, three. Thank you. I'm Vice Chairperson of Pycon JP Association. We launched the Pycon JP in 2011. At that time, the Python community in Japan was so small. The first Pycon JP had about 150 participants. Now, Pycon JP has grown to 1,000 participants over. Yes. In this year, Pycon JP will be held as an in-person event. Please come to Japan. Have a beer with me. Apart from Pycon JP event, we also worked to spread Python in Japan. One is Python Boot Camp due to events for beginners all over Japan. We have had 44 camps with about 800 participants. By the way, I introduced it at the EuroPython poster session in 2019. Does anyone remember me? Oh, thanks. Yeah. Another one is PyLady Scalaman. The event connects local women's communities across Japan. However, as you all know, COVID-19 has changed the world. We are no longer able to organize in-person events. But we must not stop our activities spread to Python. We have started two new activities. The first is the Pycon JP TV. We stream YouTube live every month. We talk about Python news, conference overview, and Python news features, and something. It's selfie time. I will present EuroPython at the next Pycon JP TV. I'd like to take pictures with you and share it on YouTube live on Twitter. Are you OK? Whisky. Whisky. Nice. OK. Thank you. I will share. I will be shared. OK. Are they? Sorry, sorry, sorry. Another one is Python Charity Talks. We wondered if there was a way to combine our spreading Python in Japan and with donating to the PSF. It is the Python Charity Talks event. It is a half-day online event with all participation and the sponsorship fees donated to the PSF. We had three events and donated $25,000 over. So huge. Yes. Thank you. In recognition of these activities, board members of Pycon JP Association, including me, won the PSF Community Service Award. Oh, yeah. It was a great pleasure for us. Please check this. Finally, I'm sorry, but I have to say goodbye because I'm now heading to Denmark because of the Lego house and the Lego round waiting for me. Super excited. Thank you. See you again at the Pycon JP annual. Thank you. Duke? No, no, you connect your laptop. Yeah, yeah. She's going to announce that we are trying to stream things. Yeah, perfect cooperation. Thank you. Do you want to borrow a laptop? Someone just said that I can do a one-minute licensing talk, so I'm just going to announce a conference that is called Pyjamas Con. So Pyjamas, like the thing you wear when you go to sleep and then Con, CO, NF, you can Google it online. I have loads of lies. So if you Google it, you will still see our website from last year. That is actually, you know, we haven't worked on it yet for this year because people who are involved in Pyjamas Con, most of them are still working for EuroPython, so we have no time to do anything. But I will start working on it probably next month, hopefully. So please search for us again next month and maybe CFP will be open. The only thing I'm sure is that that conference is going to be online. So you don't have to wake up super tired and go to the conference venue because there's no venue. And also you don't have to worry about no space to take a nap because you can take a nap in your bed. So that's it. Thank you, Chik. That was exactly one minute. We don't know drinking Guinness. We are like senior Guinness drinkers by now. Yeah. It's kind of, I'm not professional. Don't worry. I know my stuff. Okay. Hello. I'm Melan Lesnek. I'm from Kibikom. I'm a software over engineer and I want to start with the beginning, the video story. I have to come here like a first from the Kibikom and I have to wait for them for six hours. So like every normal person I waited in the pub and right down I came there or the Guinness and start checking it. We are from Czech Republic, so Czech style. Top down. And the bartender just stopped me. It was the hell on, in the pub it was extreme. He just went for me like, I, I'm mental. What are you doing? Did your mama drop you on the head? And then he told me something like when he come down that you shouldn't drink a Guinness immediately. And the problem is that the Guinness is a nitrogen-enhanced beer which is delivered in the glass which is property with the drug. As you can see, the bubbles are dragged by the glass and the current is in the middle of the glass. So the problematic part is that the glass contains much more air and the beer is not tasteful from the beginning. So we concluded a lot of research in the morning. It's going to get better, I promise. Also my pride in the pub, the guy is ugly on me and I'm from Czech Republic and our consumption is 135 liters per person in the year including infants and elderly. And I fuck up drinking beer. So research. Every day we are here, we are researching the best what we can do about Guinness drinking, what is the best average time of having to start to drink, how to look the best foam and everything. Also, we created this beautiful graph about bartender cling intent and Guinness taste. On the average, it's enough to just wait 10 seconds and the Guinness is completely fine. Sometimes it happens that we have to wait 20 seconds but it's the worst pub and you can find out by the price of the Guinness that the Guinness is like nine euros. So yeah, bartender does not take care anymore in like nine seconds because he has much more work than just take care of for you. The solution. We wanted to trade machine learning application like every other to check out the foam, thickness, color, the average priced quality of this Guinness, how much you have to wait and everything like that. So for this purpose, we already prepared a roadmap about the research preparation which happened on the Tuesday because I was here alone, so it wasn't proper research. When my colleagues came and they are awesome, everyone helped, every one of them helped a lot. Evening research, it was awesome. Another evening research. Another evening research will happen today. It is planned on the roadmap and we are searching for the investors today on this presentation. On the Saturday, we are going to research more Sunday also, Monday also and on Tuesday, we have to return back and unfortunately in the Czech Republic, we don't have a good Guinness. So cal for investors. Current total Guinness cost was 488 euros and we are going to need approximately 1000 euros and burst of aspirants. I think 1000 is enough if we wanted to have six beers, everyone, I think. Yeah, yeah, 1000 euros, it's fine. And for the investment, you are going to get whole company, whole research, everything, I promise. So contact me on my business opportunity, social medias. You can choose any one of them, I use all of them, except they become email. And let's see. That's it, thank you very much. Thank you. Sounds like a really interesting company to invest in. Yeah, I'm sorry, it was a little bit product placement, I know it's not, it's prohibited. Yeah, it's okay, it's okay. You can, it's legal in Dublin. Thank you. Yeah, this one, yes. Yeah, perfect. Cool, yeah, yeah, go. Hi everyone, I'm Omar and I'm going to talk about ML. So who here likes ML? Yeah, a couple of people. So probably in the last couple of weeks, you've seen something like this. This is called Dalimini. It's an online tool that you can write anything such as Lamas visiting Dublin and you will create images. So these are fully generated. And now you might be wondering like, where are you going to talk about machine learning? And the answer is no, I'm not going to talk about machine learning, it's Friday, I'm going to talk about memes and Lamas. And maybe a bit of machine learning as well, just to make this static. So this is a Pikachu that looks like a bug. These are Lamas participating in a hackathon in Dublin. This is Iron Man, but it's a llama. And now you might be saying this is probably magic. What if I told you this is not magic? So right now I'm going to tell you a new kind of model architecture called diffusion models in an elevator pitch. And at the right you can see some Lamas having a rave in an elevator as well. So you say Voldemort and a llama taking a selfie. There is Voldemort, there is the llama and the selfie. So what's going on? So you have the picture, right? And pretty much these diffusion models, what they do is they get noise. You have to turn the picture until you have just a bunch of noise. And then they will try to recreate the original image from the noise. A lot of text. So you say something, Voldemort and a llama taking a selfie. And then you can use a model called Clip, like Clippy, to create something called the image encoding. And it sounds really fancy, but it's just a bunch of numbers, so an array of numbers that has a semantic representation or a representation of the text. And then using diffusion models you will convert that to the last image at the right. So that's pretty much what diffusion models do. And that's how you can create a sketch of a conspiracy theory that Lamas built the pyramids in Egypt. Darth Vader moving the lawn. Among us by the sauce is a llama. Lamas in the matrix. And Lamas and Minos attending a Python conference. So thanks everyone. This demo is open source. Anyone can go on the web and try it out. And we're having a hackathon for the next week. And you can join. Yeah, thanks everyone. Thank you very much. So Christian is next. While Christian is connecting the laptop, Pycon UK is going to do a short announcement. You can use this one. So, what with me? Here. Hello, you're a Python. After two years, we're returning to Cardiff in Wales from Friday the 16th till Sunday the 18th of September. Ticket sales are open. Hope to see you there. Thank you very much. Thank you. So happy to see all the conference. There is a few lining talks that we don't have the speakers. So if you're not here, we are skipping you. And yes, looks like you have a mirror. Come on. Yeah. Perfect. Cool. Yeah, it is funny how some Python conferences try to kind of crystallize around a certain topic. And this year, I think it was very much, there was a lot talk about statically checking your code for metting your code and also about type checking. And John Carmich said who doesn't know who John Carmich is? Show of hands. Who doesn't know what quake and doom is? Okay. John Carmich and John Romero actually are the guys behind quake and doom. Well, that's for context. The most important thing I have done is a program in recent years is to aggressively pursue static code analysis. Even more valuable than the hundreds of serious bugs I have prevented with it is a change in mindset about the way I view software reliability and code quality. And I think coming from John Carmich, well, he talks about C and C++ code there. But I think the shame is applicable to Python. And John Carmich said it's good enough for me. Anyway, I held a talk around this also and I just wanted to say my slides up on the this URL and I linked I have a whole link list to that. To refer to the tools I only could I could not really address it in my talk. While I'm standing here, I want to say a big, great thank you of the organizers of this PyCon. We came to talk about what you like like our angels in yellow. So thanks a lot, guys. You were brilliant. And I like to do another shout out and that is to all of you guys to the Python community. You are so warm and welcoming and you made it so so sad that we have to say goodbye. Oh, that is the wrong meme. We'll be back in autumn to PyCon Ireland in Dublin. Thanks a lot. Thank you. Okay, so we have an announcement here. Good afternoon, everyone. My name is Par Minda. I just want to reiterate the last speaker's comments about EuroPython and how it's been so warm and welcoming. This is my first conference. Just a quick announcement. There will be a social event happening tonight about Hopper 7 and there's a small group going out for drinks and dinner, so if you'd like to join us please meet us outside the convention center around 7.30. All right, thank you. Hope to see you there. Thank you. Are you ready? After trying for two days you have a lightning talk. Finally. Okay. Thank you. My name is Konstantin. I call me Kostya. And I'm going to talk about interviews, tech interviews to be specific. And this is another talk of like five hours which is five minutes so fast in the old girls. Sorry. And originally the talk was for interviewers, but I mean if you're a candidate it's kind of hard to listen to advisors to interviewers so I decided to cover kind of both. But first and first kind of disclaimer. So I'm quite opinionated and I'm doing a lot of interviews in Kiwi and Kiwi is large. We have 200 Pythonistas or so and if you're going to apply it's likely that I'm going to be one of your tech interviewers, but other people have their opinions and I try to convince them but they try to convince me and this is one of the reasons why I'm doing this now so you can approach me and convince me back. But anyway I'm doing this talk to convince you too and I think it's going to work. So what's the problem with interviews? Basically you have only a few minutes, a few hours maybe a few days to figure out whatever the person is okay and you can work with them and you're going to happy working with them. But how can you do that? It's pretty unrealistic I would say. So what are shortcuts that people are taking? So sometimes they just assume that if you follow for all of the newest Python change log changes then you're going to know things. But I doubt it. There are people who just follow the same Twitter accounts and memes and it just doesn't work in the end. It doesn't show that they actually know something or can do the work. And the other thing is when sometimes people have a lot of questions that are impenetrable, like you can't pass these interviews and this is somehow a goal of the interview and it doesn't work as well. Of course if a candidate is smart they will laugh at you and if they're really smart they will find the question that you don't know about and they will be like, okay so I know what you're doing. So what we are thinking about usually on the interviews and technically it's universal advice that you can do with anyone who are anxious or in a nervous situation make sure they understand what's going on. Be transparent about who you are, what's the structure of what's going to happen and be clear about everything and be transparent with issues that they might have during the work and etc. So our structure of the technical interviews is like this. So first there is a kind of soft part where there is some conversation which is going on and we want to just figure out what are the interests what the person is having in their mind at the moment and during the technical interview we have kind of a quiz then we ask to share an opinion and finally we do some difficult task. So the quiz is just a five minutes to just clearly verify that person actually talks truth because I mean if you ask questions on the interviews sometimes a person can do this like Google the answer with their phone under the camera. And with the quiz you can't usually do that it's just fast but it's kind of a ping-pong question should be really simple to answer but not too hard like don't go too deep it should be fun then the opinions part well it's about to figure out if the person can learn so like people have different opinions, some people have opinions that not much yours. What we want to learn from this is how people came to this what they have learned to get with this opinion but in general unfortunately or happily interviewer have to know a lot of things so I kind of feel like this in this regard so I can't really help you here you need to know things to actually test whatever candidate knows technically what they should know but there are a few tips we can give so first if it's a challenge we for junior you better know the task really well and be able to help the junior really well so everything goes smooth then with a senior well it's better if you don't know the task and in this case we kind of work together and this feels like the majority thank you okay so the next one while you connect your computer Picon APAC it's going to do a short announcement Hello folks my name is Iqbal I've been sent over by our compatriots from in Taiwan to send you a very special and a very important message I hope I'm my mission is to invite you to Picon APAC which is going to be done online on September 3 and 4 there are two important things that we're trying to do within Picon APAC number one is a young, inspired session where school children from middle school to high school will present their projects and number two is Spring Talks on this new platform called Gather Town so I hope you can look online for Picon APAC September 3 September 4 waiting to see you folks join us there thank you thank you very much and the name of this line is Trump's Twitter yes, hey everyone I'm Peter from a small start up in Hungary called Vizu and I thought to share a story with you about a former president of the United States who I assume is the leader in this round specifically about his tweeting habits so he started tweeting in May 2009 and actually in the first couple of years he wasn't such a big user of the platform so basically his top month is like 36 tweets whereas in the years afterwards he got hooked on quite well and then he got to be a nominee for becoming the president as you can see the closer the election day comes the more they took away his phone it seems so it has a downward trend and then he won that election and became the president and there are these you know bubbles of growing number of tweets over there so these are almost all the tweets I have to admit these are only for May 2020 and you know so six months is missing before and it's out of the platform and yeah so we're going to have a glitch now and it's going to come back in a second it's going to be okay just one second not so much but yeah here we are right and let's get back okay so the interesting thing is like these were the number of tweets and this is the number of times these tweets were shared so you can say no one really cared about the guy before he became a serious nominee and then when he became president we call this graphic the Trump Twitter trumpet wow thanks okay so the number of tweets for now there are obviously three types of tweets original tweets retweets and replies and if you check the trends there are some interesting things here so in the bottom you can see the replies that's a function he kind of forget of after he became president whereas there is a huge number of retweets towards the end of this period so basically this huge growth in the number of tweets comes mainly from just retweeting things and obviously he wasn't the only one using the account so he put in the minimal effort let's just say there it's even more apparent if you see the share of the tweets a long time and then another aspect you can take a look at is like what tool he used to send these tweets first in the first period he was overwhelmingly using the web interface and then he switched to his favorite android phone and then when he became president he was banned from it because of security reasons he chose he had to use the iPhone exclusively as you can see if we take a look at the number of tweets in total sent from these platforms you can see the iPhone is ahead by far just another aspect you can take a look at at this dataset is the time of day Trump sent these tweets so it's like a 24 hour clock here AM is on the right PM is on the left as you can see there are these famous tweet trends in the early morning hours like 78 between 215 and 230 in the morning and if we filter the results based on the periods that we see in the beginning so when Trump was still a businessman he was using the web platform throughout the day and there were these tweet trends from his android phone like typing in the middle of the night when he got angry when he saw something on Fox News or God forbid CNN and when he became president these early morning and night sessions became less apparent and actually if you check the peak so when he was a businessman let me just go back the peak is around the afternoon so around here 4 in the morning whereas when he became president the peak went to the early morning hours so around 9 AM so that was the story that I wanted to share but there's a catch here because actually all I used here is Jupiter notebook to do all this this is like 400 lines of python code within a notebook I put this story together during the conference and you can do that too by visiting ipyvisu.com there are two open source packages ipyvisu and ipyvisu story with which you can build animated data stories yourself thank you very much for your attention thank you can I take a quick picture with the audience? I will disconnect your laptop so last lightning talk of today are our friends for the porto Pycom Portugal yeah you're welcome sorry so this is the last one please don't leave a stay, we have the closing session after this so we didn't use the one minute opportunity because we wanted to share with you what all of you will be missing if you don't come to porto this September we will be hosting jungle con Europe and the very first python Portugal we hope to share with all of you our beers I don't know if they are better or not but they are good enough and they are cheaper Portugal is a country that is not that expensive so even if you have already come to Europe python you probably still have budget for you and to leave you with something you can all apply for free tickets in this QR code hope to see you in Portugal thank you, I lost promising okay that was the last one in a few minutes we are going to do the closing session and then we can all go and cry in the ferryman see you in a minute