Feb12- Live Virtual Mock Interview To Real Interview For Data Scientist- Hired By iNeuron-Commerce

okay uh so once you i am sending the url in the chat so that you can check it out and probably we are live we'll wait for some time till we join until everybody joins and then we can start i'll just give you a confirmation till then i'll share in all my groups okay now people have started coming up um okay so hello everyone we have with us uh sudanshu and obviously is the host we'll be taking the interview for today's session and we have also with us smith uh so smith uh is also like that he's basically attending this today's live virtual market review uh for various roles itself with respect to data science probably machine learning and all uh so what we are going to do is that we are basically going to take up this particular interview session and again thank you sudhanshu for joining us once again and you can unmute yourself now you can also unmute smith uh and without wasting any time we'll start uh so again welcome to smith okay yes please proceed i think i've also sent you the resume yeah yeah i have a zoomer yeah yeah okay okay fine uh so uh fine hello everyone and uh welcome back to this mock interview now so let's uh start talking about so smith first of all so yeah so smith if you can talk about yourself a little bit so that we can get an idea although i have your resume so we will try to go through your resume and whatever you have mentioned we already like when do you resume so whatever you have mentioned in your resume so we will try to go ahead with the same thing yeah yeah so fine hello guys my name is sushma bengali so i have the question from commerce and currently i mean hurry up bsc-it so i got into programming right from 11 standard in python and the reason was a simple google search programming language for data science so that started my journey of python and then i got into various technologies such as android app development and question then came firebase with firebase came no sql then mongodb flutter then django and so i'm like into data science and software development boards okay that's great so smith and i think i was just going through your resume and i can see that you won multiple hackathons multiple workshops you have attended and uh achievements wise even i can see uh like uh rank one achieved in maharas uh maharashtra itg uh like a genius 2017 ms excel 2013 industry level and i can't receive jv kodi marathon and then first rank in dot tech 2018 coding uh code keita head of code gita 2019 and then it is it is not ending i guess so many things is actually yeah so many things like achievement wise hackathon work survives i can see and what i feel over here is like he's kind of a self-motivated person and he has done uh like all of those things with his own interest right so i think it will be like amazing even for us to interact with a person like you and uh take a mock interview so thank you so much for like coming to this particular platform and uh okay so i think we can start right so smith yeah okay fine so don't feel nervous first of all right so don't feel nervous uh so it's just a mock interview and whatever you have mentioned so we'll try to ask questions just based out of that one first thing second thing is that if you feel that okay fine so you should go for a google search go follow google search immediately there is no uh like uh uh it's not a bad thing right if you if you are going ahead with the google search we all know that when we do a development we do that yeah so like uh it's completely applicable so don't uh feel uh or don't hesitate uh that okay fine i'm not supposed to use a google search so just be calm cool do whatever you like right yeah and also not sure if possible you can also share the resume in the screen you know because this resume looks also very good you know yeah he like looks good and uh let me share uh my right so that everyone will get to know about it so yes guys uh this is uh sumit's resume night and he has mentioned like a spider scientist as he has mentioned that he's in a final third year right uh third year and he is from a bscit background and again so you can see all those educations detail in this particular area and uh he has mentioned over here like a professional data science with a focus on feature engineering data proposing creating a skill and custom transformations he has mentioned in this side uh his skill set right so circuit learn data processing python dart java android flutter dango sql nosql firebase rest api design tableau business analytics we were scrapping him and then he has mentioned this work experiences plus whatever certifications he has done so far he has mentioned it over here whatever hackathons that he has attended so he has mentioned it over here as well as like uh achievements and then his uh personal uh like uh attacks over here right that he has uh mentioned so yeah i think this uh resume like and he has mentioned his uh github repository link his hackers rank uh like a url and then a stack overflow summit he has mentioned over there so that is like uh everything in a single page itself and for sure even when we used to take our interview we always expects uh this kind of a resume right because it's easier for us right everything you are able to see and uh every sort of informations we all are able to get just in a single page and uh we don't have to like uh take a much headache to read out on going when when we used to go through a resume right so yeah everything is like uh consolidated in a single place itself so fine i think now we can start chris right yeah yeah yeah sumit uh basic question start with yourself tell about yourself in one or two minutes to start with yes okay so like i already mentioned i started with python so my focus regarding data science so my focus is always on like insulin modeling the first dirty work like cleaning the data reprocessing then i focus on using scalars pipeline as much as i can to basically avoid data leakage and when i found out that we had to avoid data leakage then it was very necessary for me to learn skeletons questionable transformers smith what is data leakage uh datacase is basically a concept like say you right your data you split into 80 percent 20. now when you perform some like feature engine like say you do target encoding okay so while doing target encoding if you take the data from the feed from the observations which are going to be in the testing dataset so basically you are getting data from the future and based on the data you're creating your model and then you're trying to tell your model okay predict the future now then i will paint the standard scalar on the 80 percent of data and then only i will use the standard scalar transform method on the 20 percent of data else the 20 percent of data will contribute to the mean and the deviation so you're saying that first of all you'll apply standards you'll initialize some standard scalar library you will initialize that you'll do fit you'll do fit find underscore transform and then in the test data you will do basically transform yeah what happens if you do fit underscore transform in testing data because many people have this confusion so i'm asking this uh underscore testing data then basically there is no use of cleaning it fitting it on the training data because when you do feed underscore transform all the the meaning of standard deviation will really initiate they will get overwritten so there is no use of okay so that yeah okay so uh uh fine uh submit so what we can do is so i think you must be having some id with you right yeah yeah i have kept deleting and uh as you have mentioned about the api right so first of all i would like to ask some question with respect to an api you have already mentioned like a rest api design and all those things right in your resume so uh what what do you understand by your api by the way and that too with some sort of examples which i will be able to correlate and then what in all frameworks you have used so far right in python to build an api and then maybe we can try to get into a real-time coding right so uh just first of all like a theoretical answer and then we can get into a practical one yeah okay so like what i understand and there's a server now they need to talk to each other to get instant okay so like or you can say like uh sell more or less 10 stores or a ground floor and then all 10 shows have one door so the client which has to go to one store will only go to the stores door so basically is accessing the api endpoint exposed by the store to get the required data okay so like what is your understanding about uh and what is the difference between get and post by the way so get request returns like uh in the get request you have all the parameters that you pass are in the like url you are encoded the limit is 255 and for the first request you just wrap it in the json and send it okay get the question generally used for html request so getting whenever you visit the website you find a get and whenever you request to communicate certain data say you log in on the page so you send your api key as a header the username and password to the server and the posting [Music] so you can find json okay okay so what is the uh which in all frameworks that you have used so far uh so far i have used djangolish framework i have a slight experience in fashion okay plus django right yeah very slight experience in fashion yeah django okay fine so if i'll ask you to like uh create some api and just try to expose it and show me a testing so which framework you would prefer to use whether it will be a flask or django uh for like for gesture demo purpose like we will generally use plugs okay okay but for like if you're creating a web application they'll be like a strong web application they use django okay that's great so fine uh so smith so what we can do now is so you can try to write one method right any kind of a method and then try to uh call that particular method through an api so is it possible to do and you can do it show it to us okay yeah i mean okay you can you can share your screen and uh you can start working on that just create one method and just try to call that particular method through an api and like as you have mentioned that for a demo purpose so you would rather uh prefer to use flask and for application so you would rather prefer to use like a django so fine feel free to like use any of these two i don't have any issue flash or flash be it and uh just use it yeah sure wait a minute let me just share my skin i open jupiter expecting data science questions yeah sure is my screen visible okay so yes your screen is visible yeah shhh don't know what so oh i guess it's fine let me try to learn it my import name okay i will share my full skills i cannot show the output okay so maybe you can try to reshare again so there is option yeah yeah and then share the uh just for the first option so in this way you will be able to share the full screen yeah okay so this is uh and my survey is running here on board 5000 let me open up that oh okay so i'm doing a it so let me do all my more getting questions so okay so [Music] okay okay so same same api points and no difference [Music] why um can i use concrete should i use yeah just try to provide maybe like a [Music] [Music] oh yeah here let me just get okay so this is a get request okay okay should i show you the pushy quest yeah sure good pick free right um me i guess jungle okay so i instead of slash because django gives an error because the default is equal to school yeah yeah so that's why i open this slash and then look so here's the output okay so that's great chris are you able to see it yes yeah i was just looking at it you know so some of the mistakes it's okay i think we can he can he could add or even search in the google end yeah like he is quick and uh again so in terms of resolving uh error so he's is like a a pretty good like no one remembers like any like a syntaxes and commands right and is is uh just quick so yes uh thank you so smith thank you so much yeah so you can stop sharing uh your screen and then we can uh probably move into a machine learning data science part yeah okay okay and uh that's really amazing right so you are like a still you are studying but um you are able to like uh do these sort of things and it's not like you are doing some uh btec or mtech or something so you are into a bsc it but still because we can see a hard work right the kind of uh like a hardware kind of a like a fluency that you have so we can we can see it very clearly over here right and kind of a dedication so yeah that is like amazing so uh fine chris maybe you can start with uh some of the topic of statistics right and then we can try to talk about some ml so as per a smith understanding so smith uh whatever he is is going to like a name yeah so first question what is the difference between standardization and normalization okay so in standardization basically what we do is like we for each data point we subtract it by the mean of all the data points and divide by the standard deviation okay so that's that and now by default by default what will happen what will be the mean equal to and what will be the standard like standard normal distribution the standard normal distribution with mean zero and variance one okay okay and normalizations like we can divide it by the maximum value so like which is generally used in cnn like each pixel can range from 0 to 255 or 1 255 so you're saying in case of images right yeah for images each pixel value can range from one to 255 for each rgb if we divide by 255 it will come into the range of zero to one so that's where normalization is used so probably if i have you heard of min max scalar yeah so what scaling technique do you like to call that mid max scalar so say the say our data points 10 minimum is five maximum is 15 so then minus five will be five and fifteen minus maximum fine 15 minus 5 10 so it will be 5 by 1.5 so that doesn't point to any case like sorry min max scalar what is the range uh what does it uh like the min max scale is generate from i believe zero to one of course zero to one zero to one so what do you call that thing whether it can be called as normalization normalization yeah it can be yeah so it is called as normalization right so normalization is just not only applied in computer vision or images as such it can be applied in any kind of data you have to specify the range okay what range you want whether you want between zero to five zero okay tell me something about uh central limit theorem okay so central material says that say your population then you don't know it's distributed distribution but if you take sufficiently large amount of and samples okay then you take the mean and you do it repeatedly then the the means of the samples will be normally distributed normal distributed yeah okay perfect uh that is uh absolutely right uh but i think you are also missing something right uh what happens okay from a population uh if you really want to apply central limit theorem what is the sample size you will be taking minimum 30 at least 30 okay now what will happen if you try to calculate the mean of that particular sample what will be the relation between the mean of that particular sample and mean of the population so it will like it will represent the population mean it will represent the population mean you're saying like yeah the estimate of the population it will will it be exactly or no it it it won't be exact because it's a sample basically basically say in this scenario it should be approximately equal approximately yeah into that volume okay fine uh what about uh chevinov's inequality symmetry inequality or i don't remember that it is about that okay i heard it okay uh yes you can continue yeah so uh like uh so smith can you please tell me like uh in machine learning or maybe into a deep learning right so which algorithm or which part is your favorite one uh like none of my the algorithms are my favorite algorithm changes according to the tax at hand okay so i will like i prepared this one so i will like take full four to five models i may perform cross-validation okay and then on the top three i will do hyper parameter optimization and maybe get to create and symbol of them and then whichever gives the best performance is my favorite model for that task okay that's great that is a smart answer i'll say yes you are the first one who has given me this answer right because everybody talk about one or two algorithm like this yeah yeah yeah okay so that's great okay so uh yeah uh like like which which part uh okay so what an algorithm that you have heard of uh so far and uh yeah in classification of course random foliage then svm large integration russian nervous bicycle engine gaussian name is classified okay then integration like linearization uh lasso rich electric basically a combination of those uh then clustering uh db scan uh main shift questioning amines and in classification also k n okay fine so as you were talking about dbs scan right okay so can you please uh tell me like how db scan is like a different from your k-mean clustering algorithm so how these two algorithms are different in which case i am supposed to go ahead with the dba scan in which case which situation of the data so i'm supposed to like uh go ahead with the k-mean clustering algorithm okay so like say yeah uh okay so i will just focus on some concepts like what k means does like k-means plus plus you can say it initializes in a smart way that's androids okay then it finds the distance for each data point finds the closest centroid okay it assigns that centroid it says that data point doesn't draw value so say why not do a click then it takes the mean for each question it shifts the mean like the next android is the mean of all the data points and does it repeatedly again until none of the data points extra value changes okay so okay so the drawback of this this you have to know the initial number of clusters okay because you can find them by elbow meter but you have to know to randomly initialize the centroids okay now what what okay so smith so like you said uh we had to like uh we had to have like a prior idea about a number of clusters in case of k mean right yeah a value of k so is there any like a way by which i will be able to automate this procedure maybe bus so you can use the elbow method okay fine and have you heard of like any locator uh no knee locator is another function in cyclic learn so have you heard of that one no okay fine so based on the album method so you can try to use any locator which will be able to give you like a number of clusters directly so you don't have to like uh plot the graph and look into it i think he probably may have heard of elbow method instead of yeah yeah so elbow is like a base like uh by default you look at this calling album internally so like it will be so like it checks for the variation in the likes with respect to wcss and number of clusters so yeah but yeah numerator is a function as you have mentioned cyclone that's reason so i thought of asking this question so finally now let's talk about dbs scan so i think we were talking about uh differences here yeah okay so now what db scan does say they are end points okay so they miss can we like randomly visit one point okay it will check whether it is only visited or not okay if this is not visited now basically there are two hyper parameters you can see okay one is the radius now what is the radius so radius is like it will come from far to the two dimensional data it will perform a circular boundary around it like it will create a circle whose center will be the data point you just okay say of radius five centimeter then maybe minimum number of points so now what is the minimum number of points or minimum number of neighbors so say you are standing at some point and your radius is say 5 r i is 5 so you have 5 centimeter of radius around you so 10 centimeter diameter okay now say your minimum number of neighbors is like fourth okay if you have four neighbors within that your circle okay then all of you will be caught as a clusters as a starting point you will group those it will be like a core point you can say yeah you can group those okay okay then for each data point this newly formed neighborhood you already paid the shapes okay okay now if you don't have that many neighbors then basically you and your you are an outlet okay fine if you're not able to fulfill those two criteria like a radio station number of neighbors criteria if you're not able to fulfill so in that case that data will be considered as a like if if some data will not come inside that one right otherwise if the data which is already coming in a part of the data which is already a part of core point so that will be a part of cluster now right yeah yeah and they will repeat this point for all the newly new members in that neighborhood so the benefit of this is like first you get the outliers the second benefit is like since uh you we see the number of neighbors in our area so basically uh if the density is small then of course there will be more points so you will satisfy the minimum number of neighbors criteria yes okay so we don't have to know the number of clusters initially okay and the like this is also true for main shift clustering but in mainshift flash turning uh every outlier will also make us come under some point of pressure minship questioning like it was like k-means then means shift clustering mean shape testing also depends on the insulation so the window moves towards the denser legions so that is a question but the problem is like any it will classify some point at some part of the specially it will not classify outlets rapidly this is an outlet so this advantage is in db scan yeah that's true okay so that's uh like a greater explanation about a dba scan that i have received so far from anyone so just one more thing i think he has done self-study you know with respect to data and that is what i'm seeing people are some other people i think uh sumit smith you are you have done self-study in data science yeah completely self-study right yeah and the way i'm explaining is hell name because i call my friends and explain them whether they want to listen or not so that's amazing answer right that's great i do the same thing from it so like whether people are listening or not but i do explain them and then make them like okay fine so they're like uh listening to this one okay so uh fine but i think amazingly you have explained these two algorithm okay now so anything in deep learning so smith uh only slight basic of cnn but i am mostly focused on tabular data so mostly machine learning okay so mostly machine learning and we are trying to like explore this cnn part okay fine so like uh coming to this uh classification so can we talk about our naive bias i think you have mentioned that right so how does this work and what is a mathematical logic behind that what if if i have a categorical data continuous data so yes how naive bias is going to behave so can you just talk about that okay so nearby classifier basically depends upon the new bias theorem that is the conditional probability theorem probability of a and b is like probability of big u and a multiplied by probability of a fully divided by probability of b so what is this is this so based on previous observations say uh can i actually share my skin and explanation for example proceed okay can you see the screen yeah i can see your screen yeah okay so say uh there are some features like f1 f2 and the target i think uh your screen is not visible is it visible yes uh i think it's just untitled uh this bar is uh visible so i don't know like maybe he has to shoot just try to share the entire screen okay so say we have feature one feature two and a target okay say feature one both are categorical values so say future one can take the values of a these are observations say male or female a b and say eon and feature two can take like c b i don't know see and we have some targets say pass pass oh i will kill this one and pass okay and then say we have to classify an unknown data point say where a is equal sorry uh f1 is equal to a and f2 is like c okay we have to be the target okay so what so new neo-based classifiers basically depend on the neighbor's theorem okay so like this is the initial data so now this is what probability so first we calculate the probability like the target is pass given these things okay so what is this so this is like probability so if i apply the numbers theorem which is like your point a given b so forever big given a so probability of a let's say probability of a given this is pass multiplied by probability of this is f1 is equal to n this is f2 is equal to c given the s passed uh let's call it fully divided by um okay so this multiplied by probability of a so what is a so probability of a like for all observations like how many times probability of f1 being a you can say that so probability of f1 being a multiplied by probability of f 2 being c divided by probability of b so what is probability of b probability of passing for all the levels so it is like probability that target is equal to pass okay so now probability of f1 being a given that the target is passed so there are like two records where we have the school has passed and everywhere there is eight so f4 is a so it will like two by two so it is like one so this will be one so i write one then for where they have two is c given yes pass so shall i fail this so it will be let it be so probability that f 2 is c given that the student has passed so like there are two records where the target is passed and two out of two they call the shooter has passed weight is c so 2 by 2 is again 1 then probability where f 1 is equal to a so that is number of records where f is a divided by total number of records of f1 so it is like two by three and however you have to see so it is again two by three and then fully divided by probability pass like two by three so now after computing everything this we will get the probability of like not the probability we get a value for pass then we get the value for fail then we just have to normalize the probabilities so like pass plus fail uh say pass divided by pass plus field so after normalizing it so with the help of normalization basically it will compress into zero to one the range of probability so whichever is scattered that's the answer okay i mean like i am impressed definitely this coolly i can also can't explain you know he is he's he's making all the records right for the experience like how many number of interviews i've seen i think he's the best fresher yeah so he has sure i guess he's still calling the best interview that we have taken so far right i'm not talking about all the interviews just our best interview that we have taken i think he's trying to break all of those recorders just as a fresher who has not done any kind of courses from anywhere again from just a bsc it background not from a b type or something uh i think like i don't know what's i think definitely so smith yeah so you can stop sharing your screen and [Music] i think uh chris right so yeah what we can do is so we can uh confirm his uh yes right for anyone because even in iran so we have a requirement and we keep on looking for like uh uh like openings up like we we have openings in iran for sure and uh i don't think that uh we would like to like uh lose let him go you know so smith so you have the offer with you you know i think uh you will be the part of i neuron now uh probably initially i think you'll start with internship for yours you know yeah like paid one i can say like a paid internship yeah it won't be like an unpaid one and for like a one or two month i don't think that more than one month they will be able to keep you on internship and uh after that so like uh employment or maybe you can say probation period right so more like i won't say like a intensive i would say like a probation period right and um where all the six and when when you are going to like uh complete your uh bseit uh so you're like a college like the schedule is me okay may yeah so before that i think you will not be even able to join uh right so maybe from now onward so you can start working with us right uh maybe it will be paid it will be paid internship yeah so tomorrow onwards and uh unless and until you are not going to complete your uh like uh this one exam so full time will be yeah so you will be working with ir and after that so after just once once you will complete your college from that day itself uh you can can just mention your date and from our hr portal so we'll send you uh like um uh like uh later and uh yeah so you can join just as a full-time member of anyone uh from like a day your exam is going to get completed everything will be mentioned your compensation part your salary part your perks your benefits right each and everything will be mentioned in our offer later that we are going to send it to you uh just feel free to do a negotiation on that part we don't have any kind of issue on that or whatever like uh we are going to send it to you and uh yeah and anyhow from tomorrow so join an internship and the paid one and i i don't say it's a for you it's intensive uh just join as a full-time simple even though you're working you are like uh doing your college but still just join it and then like after you call it so start like a contributing i need on as a full-time simple okay so this is my first interview like question they went for a small interview yeah first mock interview a person who has not joined any kind of uh uh like a courses just whatever is available on internet or as an open source so a guy who is like uh trying to learn from there and the way you are trying to explain the way i i can see a code right so whenever person will start writing a code and just in one or two lines so we get to know that okay fine so what is a person's capability uh in terms of writing a code in terms of explaining that things and how depth uh idea or understanding you have and and we can we can see and we can feel that how you are able to understand right how how you have understood all those things when you were talking about dbs care when you were talking about made bias when you were talking about cayman clustering right the way you were explaining right so i completely correlate like when we used to learn right by ourselves so the way we used to think the way we used to like imagine we used to visualize and then we used to correlate and uh we used to explain these things to others i think it was a amazing one and this is now one of the best interview that i think me and chris has taken so far and i don't even compare with fresher you can understand right we have offered you the job now right yeah we don't do this in live right tomorrow you'll have your offer letter everything uh probably yeah just something i'll like send your email id and uh email id you just send it to me i'll send it to uh yeah so you will get a like a mail from our hr portal right uh it will ask you to fill all the basic informations your college and all those things right your like uh normal id uh details like your pan card or if you have otherwise or something right just try to upload and complete the formalities and then i'll introduce you tomorrow with our team in that case yeah yeah so yes i think not tomorrow so saturday right saturday sorry monday monday we'll just take your uh but on voting positive will be done uh in this uh this saturday itself this saturday today so maybe like i'll ask someone to send you the hr portal and you can complete the formalities at least okay yeah just complete the formalities then we will uh we'll release the later from our site okay yeah sure today itself so we'll try to close this one okay so yes that is great so anything chris how do you feel smith say we're looking for this kind of candidates probably that is the reason we do virtual mock interview and all but again it is really really difficult to impress sudanshu if he's impressed i think he's directly offering you the job right so it's amazing uh that i think this is the first time that we have offered well we have offered job directly in the live virtual interview session like mock interview we told her but this has really become a real interview you know so amazing that he has done right so again everyone would love to like on board him uh right as a part of self as a part of any organization and the person who is a self motivated one right because we always look for like a person uh who will be able to like uh explore new things because data science is a volatile field itself is a volatile area you can't expect that every things will be taught to you and then uh like uh in everywhere everywhere like someone will do a hand holdings and then uh someone will tell you what to do when to do how to do no things never happens right we always look for people who is a self-motivated and i think you are full of motivation you are a motivation inspiration for other people as well maybe other people can take you as an example any and many people asking what role you'll be getting probably like a junior data scientist yes associate data scientist and then uh like uh for sure like he will like uh work hard and then uh as per the hierarchy of the organization so he will be moved yes anything that you have questions for us so smith regarding the like the the details i've been making you personally yeah don't worry about whatever yes anything apart from that how is the interview completely unexpected completely unexpected okay perfect uh it's okay okay fine uh uh anything cilantro before the ending you want to say i think guys i hope everybody liked this interview uh this was the first time probably in a youtube live session we are offering someone job uh probably we go to around two to three rounds to do this but uh i don't want to lose it because simple yeah we don't want to lose him so we are directly offering him over here and this is the proof you know you have this video always available in youtube so don't worry your job will not go anywhere okay yeah okay guys okay so uh guys yes this was all from our site hit like for smith for this amazing interview and whoever is coming next prepare well and come you don't know whether you'll get directly hired from here okay that is the reason i say prepare everyone prepare and come that is the main thing prepare and come prepare and come i usually say and finally so smith i think so smith was actually telling me from last week krish when is the interview when is that interview and in his interview i think he was doing it from last week itself finally today we were able to do it uh he prepared well he was able to give the interview and today he got the offer this is the life offer that we are giving him and i neuron uh initially start with internships and after may probably he'll be joining a full time after he finishes his college but yeah everything will be paid for him so yeah internship will be paid paid paid paid because he has done that all entirely self-study you know he's not been taught by anyone in sure so yes uh this was it uh thank you sudanshi once again and this is an amazing step by providing job offer so thank you again for doing this uh uh smith is speechless i guess right now his expression i can see it okay uh so you're happy to mr smith i guess yeah okay so perfect so congratulations to smith and uh hit like for sushmit guys and yes uh keep preparing keep learning you don't know you may be the next person who wants to come and interview attend the interview you may also get a job because ask you said that it does not want to leave quality people you know so i hope i'm right telling it right so then sure yeah and we have our openings we always have our openings right so again you know that uh today itself we were talking about like one new division right so like uh we are like uh working on that part as well so for sure even for that so we are we are like on boarding many members so yeah that is like uh something that we have and for sure even we have a contact so we are like uh many people and just one one more announcement if you don't know sudan show and myself are starting one more startup we'll be announcing that in the next week it is a surprise for you all and probably you will be able to get more jobs more jobs and more jobs you know and that is also related to ai field only but that is in the sector of what sector you want to say that illegals so that is the startup that we are coming up it's almost done everything is done uh next week the website will be available and all uh so it will be available there also we are going to implement ai things you know so more jobs more opportunities for everyone and that is the reason uh we are working for that only so that we will be able to create more jobs and as much as possible so yes i think i made the announcement also today only finish the discussion and uh probably next week uh probably i think so smith may work in that i don't know yeah so uh that that is a like like forward supply yeah yeah so it is a completely product-based company so yes company profile everything will be shared next week friday where i'll be making a video and introducing our startup again for this particular work so it is an amazing work a lot of ideas have come in mind like how we are going to do it and the domain is taxation and legal uh suppose you want to register your company you want to access something in the next announcement yeah yeah sorry sorry i'm a commerce student okay so i see domain expertise also come so don't need to worry about anything smith can't do anything i have like uh you see i have seen uh like uh people like so smith right so like uh this kind of people actually you don't i think if he does not know something right he will try to learn it because he has learned this much things right because that way he has actually explained he has went into depth right and that is very difficult to do it because i can understand the problem because we have been learning from many days right so let's see what will happen and yes uh we'll be announcing next week guys so thank you all uh from our side uh thank you sudanshu thank you smith and yes guys congratulations to you smith from my side from chris site from all iron team right so and many people many people in the audience is also saying congratulations they are very very happy that you have got a job so thank you thank you everyone thank you again for being a wonderful host and finally thanks all the time consistently yeah today today i was about to force you please hire him you know you told it uh so i was like okay fine you understood my thing we understood everyone's feeling right even family say the same thing okay okay guys so this is it from outside and yes we'll see all in the next uh session so thank you everyone uh so i'm sure we can be in the zoom meeting uh just be in the zoom meeting and thank you everybody uh we are stopping the live stream bye bye everyone bye hit like for sumit hit like for subscribe hit like for me finally okay i want multiple likes two times if you can do two times like do it okay

Channel: Krish Naik

Views: 116,309

Keywords: data science jobs, machine learning, deep learning, krish naik virtual live interviews

Id: Hf4qbtzxc-Q

Length: 55min 25sec (3325 seconds)

Published: Fri Feb 12 2021