Enterprise Brain Scrum - 2026/03/18 07:55 PDT - Notes by Gemini Mar 18, 2026 Enterprise Brain Scrum - Transcript 00:00:00   Rajashekar G: training calling. Manisha Gundapuneedi: Oh, training. Stop just Session ended after 00:01:48 00:16:07   Vara Kumar Jagarapu: Five Rajashekar G: Okay, Yeshwanth Reddy Yerraguntla: Manisha. Vara Kumar Jagarapu: minutes. Rajashekar G: enough. Yeshwanth Reddy Yerraguntla: But all six of us are were in that same call, right? Rajashekar G: Not we too. Vara Kumar Jagarapu: minutes. Apollo Manisha Gundapuneedi: Go Yeshwanth Reddy Yerraguntla: Mhm. Gopal Gottumukkala: time. Manisha Gundapuneedi: morning. Gopal Gottumukkala: What is the time that is good for him? Okay. Yeah. Good. Yeshwanth Reddy Yerraguntla: kickoff call. Rajashekar G: Okay. Gopal Gottumukkala: Kind of uncondition. We just want to make sure that you know um what do you say who will work Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: and plan so that we have to update Pakaro also so that he knows who is working on which product right on Jen HRX he's not a delivery manager but he still needs to know Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: who is go going out of the picture from enterprise grade so let us think what could work for us Rashak you're also part of it right so some kind of you know borrowing loaning has to happen and then finally we have to figure out it's a tricky project had a look at 00:17:35   Manisha Gundapuneedi: It's Rajashekar G: Okay. Gopal Gottumukkala: it and uh yeah we need a time also but yeah we need to finalize the team start and then see in which course we will Rajashekar G: Okay. Gopal Gottumukkala: take Satyasri Prabhakar Mantripragada: That's Manisha Gundapuneedi: 95 student. Satyasri Prabhakar Mantripragada: it. Manisha Gundapuneedi: Okay. Yeshwanth Reddy Yerraguntla: I already chose Uh yeah, Manisha Gundapuneedi: question. Yeshwanth Reddy Yerraguntla: we'll meet invite. Manisha Gundapuneedi: Huh? Okay. Yeshwanth Reddy Yerraguntla: Okay. Okay. Manisha Gundapuneedi: Uh 15 Yeshwanth Reddy Yerraguntla: Unless our scrum Vara Kumar Jagarapu: Very Manisha Gundapuneedi: minutes. Yeshwanth Reddy Yerraguntla: extends Satyasri Prabhakar Mantripragada: Enterprise. Manisha Gundapuneedi: Bye, Vara Kumar Jagarapu: good. Manisha Gundapuneedi: sir. Vara Kumar Jagarapu: You're good. Manisha Gundapuneedi: Amen. Rajashekar G: actually call discuss that is what ask I asked you today morning. Manisha Gundapuneedi: Amen. Rajashekar G: So story. Satyasri Prabhakar Mantripragada: That's Rajashekar G: So make Gopal Gottumukkala: The one good thing about that project is many of the concepts will come back here. Rajashekar G: call Gopal Gottumukkala: I I mean I don't know yay but by looking at the problem statement the kind of anticipations that they have many of those things they will be pushed for us to make 00:19:01   Satyasri Prabhakar Mantripragada: Cool. Manisha Gundapuneedi: Sorry. Gopal Gottumukkala: it right so some of these things will certainly help us in enterprise brain is what I'm thinking might be totally wrong also not the business functionality but the Yeshwanth Reddy Yerraguntla: Yeah. Especially Gopal Gottumukkala: the engineering behind that should help us Manisha Gundapuneedi: Let's Yeshwanth Reddy Yerraguntla: in the LLM sorry guardrails aspect right we will learn a lot the other two which are authoring and Gopal Gottumukkala: Yeah. Rajashekar G: Okay. Yeshwanth Reddy Yerraguntla: uh rank encoder we've never done it so we will learn that also. Manisha Gundapuneedi: go. Yeshwanth Reddy Yerraguntla: Yeah. Um so today morning whatever call we had mainly we were starting to see a few pain points that we wanted your help with goals main while we are coming up with plans in Jira we are putting them up we feel there is some miscommunication or there is a better way to do it at which and we want to pivot but how to do it properly that kind of a guidance uh we are missing like when I if I had the freedom probably I'll commit to something but in the middle of it I would have said okay this is not correct and I'll just put that aside take up something else that depends on that and execute it one example that we saw was there There's a question answering C necessary in uh the overall module right like when user asks a question 00:20:53   Yeshwanth Reddy Yerraguntla: especially on structured data LLM need not reinvent the query every time or similar questions you store the plan on how to generate the query that plan will be useful again and again that is a crucial component we assigned it to our pit But because I said it's a crucial tool for structured data, we are not building any connector right now for structured data. The only structured data source we have is Jira and we are dep prioritizing it. So even if Arpit builds something, there is nothing for him to show. So I was under the impression that we'll anyway push ourselves to still execute and deliver Jira. But I came to know that uh with the current time crunch with the number of projects we are handling we can't afford to do these things. So even though we assigned this tasks to our pit now we are realizing this is probably not what we should be doing adjustments so on a daily basis or at least every two days if we are seeing I feel bad that I'm confusing the team into doing these ad hoc things without a proper plan. 00:22:15   Yeshwanth Reddy Yerraguntla: So uh that is one very crucial reason why I wanted your guidance. Second reason I wanted your part in guidance is while while we created the task we I still am sucking at uh creating a timeline and uh creating a delivery timeline for praakers. So saying by this next two weeks we'll deliver this. These two aspects I need help Raj or Manisha you can add if there are things that I'm missing. Please Rajashekar G: Let's go one after the other like we listed down here like the Manisha Gundapuneedi: Is Yeshwanth Reddy Yerraguntla: add Rajashekar G: sheet these are like uh general points and these are technical points. Manisha Gundapuneedi: it Rajashekar G: Uh today while discussing regarding org snapshot and uh alert life cycle syncing and everything. So we came up with some questions. So those questions I listed out. Yeshwanth Reddy Yerraguntla: What's Rajashekar G: Yeah. Vara Kumar Jagarapu: observation and top Yeshwanth Reddy Yerraguntla: that? Vara Kumar Jagarapu: level. But when it goes to execution low decisions point of view we are again questioning ourselves uh to make it 00:23:36   Manisha Gundapuneedi: It's Vara Kumar Jagarapu: system. Yeshwanth Reddy Yerraguntla: What's going on? Vara Kumar Jagarapu: Yeah, that is what my Gopal Gottumukkala: Okay, so far three aspects. Anything Vara Kumar Jagarapu: observation Gopal Gottumukkala: else? So all the three are typical problems right? So see you are running too many projects so it is becoming overwhelming. If we are working in one single project also the same things will happen but the frequency will be less. It it's not like you know every day thrice but at least once or twice in a week or Yeshwanth Reddy Yerraguntla: It's all matter. Gopal Gottumukkala: minimum to minimum couple of times in a sprint that's very natural probably we are taking heat because we are taking guilt out of it. So keep that aside what are we trying to do and how are we trying to do if we can concentrate on it that helps us whether we are doing wrong or right. Right. What are we trying to do is something that we thought at a certain point of time it felt like it is easily possible but once we saw that easy part through then our dreams became big slowly comp complexity came in and then now we know how big it is so I'm not trying to scare that that's what happened right with us so slowly we our dreams become bigger and then we wanted to add more constraints to it 00:25:23   Gopal Gottumukkala: or features however you want to call it governor is is a feature but in on the system it is a constraint right that constraint itself is a feature so things like that start coming in they they're not like you know it's not a simple AI problem it is a mix of mix of both it's a general business problem has to be converted to part engineering, regular engineering and part A. So that is where you know we we have to take certain calls to prove the point we have to do something that might not be there in the priority. The classic example is preserving the query generation plan. Yeshwanth Reddy Yerraguntla: You Gopal Gottumukkala: It's a good idea in my view. Query generation plan we have to think through conceptually the plans needs to be Yeshwanth Reddy Yerraguntla: sure? Gopal Gottumukkala: preserved be it structured or nonstructured if you ask me there is nothing like unstructured in reality everything is structured so you look at look at a data science anything it has to be structured at the end uh so human brain cannot work on unstructured information that's the reason we could not able to crack medical problems yet conceptually medicine how brain works these are illogical that's the reason we are struggling to solve that problems whatever the level that we thought we solve those problems we converted them into logicals probably that logic is right is the assumption so far that's the reason the same disease you get a new medicine in another medicine 00:27:04   Gopal Gottumukkala: because our understanding is increasing that the medicine is changing. So the problem is not yet solved. The problem is solved to the level it got converted to a logic. Something else breaks the logic. We'll figure out enhancement of that logic. That that is how that's one theory. Basically guys don't take this as a literal uh you know analogy. But that is one theory and you know what is unstructured. So when we tried solving unstructured problem long ago, we got confused for few days and then we realized that there's no way you can solve unstructured problem in unstructured manner. This approach has to be structured. You problem source could be unstructured then obviously I mean we solved in a certain way we attend something that that's history. The point that I'm trying to make is that concept is very powerful of you know preserving the plan. We are trying to save few cycles that means few milliseconds accumulated you you're reducing the latency right that's a very critical part reducing latency we can do it in multiple layers of a particular transaction reducing latency by avoiding waste uh you know seems very simple from outside but solving those problems are the last problems that we usually solve if you catch threshold of such a scenario you better fix 00:28:33   Gopal Gottumukkala: it because you don't get a chance or you don't get that necessity so you will not be pushed to fix whatever reason we got that that's a good one now how to take forward Jira got uh dep prioritized now in genhrx we have a structured uh data source right ultimately that is structured in the sense how many posts I created out of which how much my rating is and what you know uh what I thought it is and what other people thought it is and out of those other people some are my followers some are outside of my followers then how the rating is these problems needs to be solved that is purely structural again see figuring out rating Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: is an AI problem that's one problem and then once that came back and became structured laterally structured right we have to store all these Yeshwanth Reddy Yerraguntla: Yeah. Gopal Gottumukkala: things there could be an a solution on top of it but it is partly engineering partly a problem it became there is a structured data which we are already presenting to the customer we cannot lie about it on top of it we have to draw correlations again a came into the picture which is nothing but your structured enterprise brain right basically know we have structured data though we know the schema we still need to figure out the question is no more SQL right question became textual same same problem 00:30:08   Gopal Gottumukkala: right small segment of that project so this this problem cannot be pushed away it is important so if we have some data set that we already fed if we want to reduce that effort there schema has been fed to some tables right can prove that point it is an add-on feature to Yeshwanth Reddy Yerraguntla: All right. Gopal Gottumukkala: I yeah being paid for that or not is a different story but that problem has to be solved. So once the problem is solved then we can think about is there any such opportunity in unstructured I don't know if there is nothing we leave it off we'll limit that you know technique of optimizing plan creation we'll limit it to the structure but if we find a way we'll apply the similar philosophy solution I mean exact implementation would be different there but conceptually the optimization aspect can be applied there also Right? If I look at it, it is a problem that was picked right and the Jira might not be the priority on this, then pick another priority which justifies this call. Or else if we are thinking to put it into the backlog, add this aspect to it. 00:31:34   Gopal Gottumukkala: If the intended data source is not available then what is the next best data source how I can create a business value right so genrx I already showed one cause IM I showed cause here it is a mandator right in genhx we already promised IM of course we promised or not Yeshwanth Reddy Yerraguntla: All right, Gopal Gottumukkala: we might not get value unless it is asked there's no point spending for them you can spend for a different reason. Yeshwanth Reddy Yerraguntla: got him. Gopal Gottumukkala: If it is purely for them because they are not asking that might be a good call. So that is how you make a call and if as long as you fix the path right now we left a small piece in the path no data source right attach a data source to it with whatever conditions that that makes sense. Then take a call. If you want to uh push it to backlog, push it to backlog or want to continue, continue. It is whether it goes to backlog or whether it goes to execution is based on the current business value that we are anticipating out of it. 00:32:47   Gopal Gottumukkala: That that's a call that we need to make. That is one thing. Whenever we are shuffling things then what happens is whatever you are promising to pragu is not set in stone because in any number of directions priorities are changing for a good reason. Okay. through the the change pushing out from scope pulling something into the scope is not just happening because you haven't thought through or somebody else is not thought through there are other reasons that is happening. So we have to keep that flexibility when there is something needs to be done today in order to get something in the next week that needs to be done. That's a logical discussion that we should have and then make a call. Push or pull should not matter at the end. We just need to talk and make a call stick with it. As long as that that you know the value that we saw that time either to pull in or push out is not changing we can stick with that decision. If it is changing rediscuss, pull it or push out. 00:34:01   Gopal Gottumukkala: So sprint is constant and the effort went in is constant. What items goes in, what items are coming in is purely business value driven or tech value driven or something you know it's going to come. As long as we have this clarity, Prabhakari should be fine with that. what comes in what goes out because it is happening on other side also right Prabhakaru got an instruction or some other sales deal pushing a different aspect that we thought three sprints down the line we are going to do could have came now and something else has to go out right at that time also same call what is the least business value or no technical impact judgment make a call pull out something pull in We we should not take guilt out of it. That's that's the nature of product that we are trying to build and it is accepted inherently either directly or indirectly. We accepted for it. So there's no point feeling bad about the decisions that we are making. I also understand there is another part. 00:35:17   Gopal Gottumukkala: Are we confusing the team? We are confusing the team if all of us doesn't have this kind of clarity, this kind of understanding that it is a dynamic day. If we know it is a dynamic day, whether X goes out or Y goes out should not matter. Right? And I already know Monday to Friday, you know, eight out of my 20 tasks for that week are going to change. Then it doesn't really matter whether it is X going out, Y going out or something else coming in cuz I already know something is going to change. If not, we have to get into that mindset. See especially any project that is very close to pre-sales that means we are trying to sell this is very natural road map is road map we have to branch out come back branch out come back we just need to have that accept acceptance level otherwise it will be very difficult uh for all of us emotionally okay if any of us are taking that heat that you My tasks are changing every day. 00:36:39   Gopal Gottumukkala: Luckily, all your team is highly mature and understanding people. As far as I see from outside, it is just about sitting on the table and talking about it. At this age, what these guys are going through is going to make them the top class debuggers. The first quality of debugger is just shut off one door, open another door and make an attempt. just instantaneously. That's what they are going through in a job. It's like a gift. Am I making sense or I'm just b***********? Rajashekar G: making sense, Gopal Gottumukkala: Oh, the first problem is make the calls. Rajashekar G: sir? Gopal Gottumukkala: At that point of time in the context that we have, we are making a call. If we are reassessing the call, something would have changed. If nothing has changed, our call is right. If something has changed and we came to know, we'll make the we change the call, right? Manisha Gundapuneedi: Yes sir. Gopal Gottumukkala: And I don't want us to think because it is technical so we can push it up. 00:37:56   Gopal Gottumukkala: No, though it is technical, we still look at the business value what it could produce either today or tomorrow. We're not doing it for sake of proving the point, right? We already did that. We don't need to prove everything. We just trying to create value. So make those calls and then update move on. The only thing that you Prabhakaru v and Manisha and Rashakra has to take care is making the team understand the value in the Dismating and continue to do. This is not a regular development, right? This is not regular engineering. So it is different and it is special. Yeah. Unfortunately, these are not promoted much in the industry because it's a painful path. Imagine somebody living almost like 17 18 years in such an environment. that that's it's a different level of mindset different level of skill set it will always do good to us at least in the 17 18 years of juggling I never felt bad that I made calls yes 90% of my calls went bad but that 10% calls which went good produced enough produce enough value that the remaining 90% nullified that's the reason I have a job otherwise I should have been thrown out long ago right I can only make a call based on what I have today I cannot make calls on imaginary stuff 00:40:03   Gopal Gottumukkala: right when the day passes on something changes don't look at that call that it's a bad call that was a call good on that day that moment mind that information. Same thing. So right now optimization is important then optimization is important. Flip the data source pick another data source. If optimization can wait, Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: it can wait. What arpit needs to understand is okay shut that door. Take five minute break move on to the next door. For that matter any of our any of us there is nothing wrong from your side there is nothing wrong from side there is nothing wrong from side or not even from pratima or sales team for them also it is our changing right every sentence customer is going to make our necessities will change we still trying to sell when this will reduce to a very very drastic Yeshwanth Reddy Yerraguntla: H Gopal Gottumukkala: level when this product is established Then there is a thicker road map aligned to what it is doing and what it is going to do. 00:41:18   Gopal Gottumukkala: Then you can always show them that this is not the path it should take. So it is very off. If you want we'll build a custom solution. We don't want to tinker with the platform. We are not there. We will reach there but till that point of time the dynamism will be there. Third aspect what V was talking about right do don't spend too much of time whether it is right or wrong again goes back to the first point we are in a dynamic environment we agreed that it is going to be dynamic we know that it is going to be dynamic so you could make wrong calls but luckily you guys are connecting on a daily basis. So your wrong step could max to max only done a day. Okay? You have people around make a call then the next stand up if something went bad take the call back and then take another approach. See unfortunately even if we spend 3 hours to make the call which you are supposed to make in first five minutes if see we missed information is a different story. 00:42:45   Gopal Gottumukkala: I'm assuming that we have all the information and we are considering that information. Whatever the call that you are going to make in 5 to 10 minutes of discussion after 3 hours also it doesn't change. No new people are coming. No new information is coming. Right? With that assumption make a call goes wrong. You're betting for a day. Right? Most of the times your calls are going to be good because you are making calls in the context. Again I'll go back to the common understanding. We don't know exactly what we are going to do after 3 months but we know what we are trying to do now. What are we trying to achieve and the way we need to work and things are changing. These things are common and well understood across the floor with these assumptions. Yeshwanth Reddy Yerraguntla: f***. Gopal Gottumukkala: This theory works. Okay. When you need to spend days and days of whiteboard is you are trying to write something in store. 00:43:53   Gopal Gottumukkala: You can't afford to buy another store. Then you have to spend time. But more or less we are trying to figure out the best path, figure out the best tool, figure out an option to fix a problem. We cannot pick that approach. We are just trying to figure out what needs to be written on the stone. We are not writing in the stone yet. If I continue with the same uh analogy, right? The way I see it, it is more of probably it is coming out of some of our calls went bad and then we thought we would have made a different call. But that is going to happen. No matter how much we protect ourselves by spending time, that's bound to happen. I'm not trying to discourage you. I'm just trying to show you the the reality. I don't know guys how many articles that you are reading, everybody is claiming that they did something similar. But I want you to concentrate on the examples that they producing. 00:45:06   Gopal Gottumukkala: Just don't go by the theory that they wrote in those that the use cases that are demonstrating right very superficial any article the use case is superficial I usually don't continue I'll just ignore that part I take the concept okay something like this is happening in the industry and then I'll come out because it it wasn't went into the depth that's the reason they don't have the use cases. I mean of course they can talk about complex use cases but if they're demonstrating then if the demonstrated use cases are pretty high level that means they also did not reach to the depth that we are trying to go. The point I'm trying to make is it will be the same life for them also for us also whoever is trying to do which is not an established uh you know Yeshwanth Reddy Yerraguntla: What are you going to do? Gopal Gottumukkala: theory everybody struggles this uh there's no magic pill that they found in Apollo and we went to med plus I don't think that is the case here. So, so far what I spoke about you or Ashwan or Manisha do you have any 00:46:41   Yeshwanth Reddy Yerraguntla: So embrace the chaos antar Gopal Gottumukkala: Yeah, that should be should have been the first sentence, Yeshwanth Reddy Yerraguntla: H Gopal Gottumukkala: but that would sounded like came from a book. So, I spoke everything else except saying Satyasri Prabhakar Mantripragada: Oh s***. Gopal Gottumukkala: that Yeshwanth Reddy Yerraguntla: but there are a few points which I mean now I from hearing these sentences I I'm what I'm about to ask I probably would have asked in a different Right now I'm asking with the uh optimism that it is still an opportunity not a problem that how can we how can we uh ensure that we are keeping our stakeholders uh um happy even though you know we are not producing anything tangible today. How can we ensure that they can be confident in our ability to deliver whether it is 2 weeks down the line or four weeks down the line in the face of these changing requirements like what is the best way to communicate uh ideally on a day-to-day basis but otherwise also that when things are changing how they are changing and communication channelity this currently where I'm thinking 00:48:09   Gopal Gottumukkala: So keep the daily status intact but make the commitment window a little longer. So in the sense what I'm trying to say is to showcase something unless and until it is very very urgent dynamically we are doing it that is fine otherwise the planned things don't plan a demo on a daily basis. How we can do is increase the window to 2 days or 3 days. See we are all working in different pieces of the puzzle right? So one puzzle don't make a daily basis. One puzzle make it create a two-day 3-day window. Second puzzle will be the next day. Third puzzle will be the third day. So that the first puzzle whoever is solving the first puzzle they get two to three days of window to demonstrate the outcome. Otherwise what is happening is every day it is something pending. Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: Next day something pending. Next day something pending. Right? See ultimately we are not working one piece. Right. Uh as far as I see uh how the things are happening, we are obviously working on more than one piece. 00:49:22   Gopal Gottumukkala: So today target one, tomorrow target another one and then repeat that cycle so that you you give enough chance to yourself, me and the person working also. I'm not saying Yeshwanth Reddy Yerraguntla: So the deliverables Gopal Gottumukkala: I'm not saying reduce the productivity, don't show demos or don't updates nothing like that. What I'm saying is this will continue but on the same aspect uh let us not demonstrate today morning, tomorrow morning, day after tomorrow morning. Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: Sometimes it might be say if we pick three bugs bugs fixed it is fine. Sometimes we picked a task so which is about anticipated to Yeshwanth Reddy Yerraguntla: Hm. Gopal Gottumukkala: be 2 to 3 days assume right in that 2 to 3 days in general engineering problems on the day one there is nothing to show except the code Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: that you wrote. Usually end of second day there is something comes ideally third day mid is somewhere you have something to show right. Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: So the outcome demonstrations if we can plan that way there is no fixed formula 1 day 2 days or 3 days but pick based on what is in the hand there is a possibility of outcome then there is a then we'll demonstrate but if we couple it very tightly to the time frame right always like you know uh every day there will be a demo or every Tuesday there will be a demo then probably we are not planning the task well rather than planning for a 00:51:19   Gopal Gottumukkala: good delivery we are planning for what to show tomorrow then obviously engineering will take a dent if not on all all the tasks quite some tasks will take a hit because Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: we plan differently right ideally if you ask me to uh build a file upload plug-in you will get an upload capability after 2 days if I have to demonstrate upload capability on the day one. It only works for 10 m I mean sorry 10 KB 25 KB. Moment you put an MV file it will fail because I have not taken care of the things that I'm supposed to take care of streaming and making sure that I create enough time delay on the server side also I pipe it rather than directly writing temporary storage hookup all these things I'll ditch right so it it feels like working but something changes in the input it breaks So those kind of tasks it doesn't make any sense to you know show the upload window on the day one for 1KB I will work for 5 MB later 25 MB later then obviously I'm already polluting my code which I'm supposed to start with protectors and streaming I started with the blunt approach I'm refactoring the code be it code generated or written that's not a good way to right code that my point right so now we have to think is there a possibility for 00:52:58   Gopal Gottumukkala: that then that will help us what do you Thank you. Yeshwanth Reddy Yerraguntla: deliverable. Satyasri Prabhakar Mantripragada: To whatever extent we have clarity still where is the communication gap? Rajashekar G: Sir my point is like milestone one we started milestone we will be demonstrating something discussion start so now when we are working in the environment like ashwantana has put up like Jira into the picture but coming to the milestone way the Jira is at P5 or Satyasri Prabhakar Mantripragada: Thank you. Rajashekar G: P4 level priority level So that needs to come out and again right now whatever we are working on the milestone what is the timeline every time end of the third week what are we going to show end of the fourth week what are we going to show and clarity still we are not able to Satyasri Prabhakar Mantripragada: I'll drive it. Rajashekar G: out. Satyasri Prabhakar Mantripragada: Uh see when I was working, we also had the same thing. But what we used to do on sprint one and two were the plan for the sprint three and four. We used to present that okay this is how we are going to do. 00:54:39   Satyasri Prabhakar Mantripragada: Sprint one. Rajashekar G: Yes. Satyasri Prabhakar Mantripragada: We'll only show them okay this is these are the highle designs. Gopal Gottumukkala: Hey. Satyasri Prabhakar Mantripragada: These are all the low lowle designs. This is the test data prepared execution starting Rajashekar G: Okay. But so today's March 18. Satyasri Prabhakar Mantripragada: started. Rajashekar G: Let's say we started on March 15th or 14th Monday. I'm expecting March end week. We have already done this part. Is there something to demo or not? Satyasri Prabhakar Mantripragada: and Rajashekar G: Uh so question like I'm checking with Satyasri Prabhakar Mantripragada: other Rajashekar G: and in the team like are we clear with the plan or not every day every day it's Satyasri Prabhakar Mantripragada: this one we already discussed. Rajashekar G: happening Satyasri Prabhakar Mantripragada: So from our standpoint okay story points velocity but management standpoint we have already started in Rajashekar G: right. Satyasri Prabhakar Mantripragada: Jan we have done some ground level work how perfect is that what are Rajashekar G: Yes. Satyasri Prabhakar Mantripragada: the what is its goal versus where it is now 00:55:49   Rajashekar G: Yes. Satyasri Prabhakar Mantripragada: okay you already started in Jan now you have almost coming you are coming close to the merch end. So anything that you build on top of the existing one you are actually adding more flavors or more layers is how they would visualize. Rajashekar G: Did I Satyasri Prabhakar Mantripragada: That's why they would ask you to demonstrate something so that they also see progress in the application and then they commit to the new clients on new capabilities of the application. Rajashekar G: All Satyasri Prabhakar Mantripragada: They cannot sell the same story to the same customers because for example consender right they have come to us Rajashekar G: right. Satyasri Prabhakar Mantripragada: in early February we have shared them that okay this is how the UI looks like and it can be integrated with Salesforce. Now if they come back and ask them with curiosity that oh now show us what do you have we have already talked to one of our clients and they are interested in it. If you again show them the same thing then they will not be interested. 00:56:49   Satyasri Prabhakar Mantripragada: They'll not we cannot sell the sell the stuff that we have built so far. So there should be so this is where the prioritization comes right you should have something that is at the core something at the front end layer something on the integration layer. So you have to move multiple gears so that you can make progress in all directions gradually. If you say we'll only complete this part then only we can go into this probably sometimes it'll be challenging for us to demonstrate that and there are certain aspects which you cannot demonstrate also for example you have finalized an architecture in sprint one you worked entirely on architecture you finalized it you cannot demonstrate anything right so but we took the time to dig deep into understanding what we wanted to build how we wanted to and we kind of if not 100% at least 80% we have nailed down the architecture that's my understanding so to me that is a clear picture that we have started with right I'm not sure how many projects in the past at least for engineering we have taken this much solid time to only work on the architecture to a decent level and then we are getting started on the sprints So that's where I kind of I was informing Naven also that no we are working on architecture we are working on architecture once that gets nailed down we'll have a clear picture 00:58:24   Satyasri Prabhakar Mantripragada: but as I said you need not build a foolproof 100% perfect Rajashekar G: Okay. Satyasri Prabhakar Mantripragada: system even if you build with 60% clarity that is good enough Rajashekar G: Right. Satyasri Prabhakar Mantripragada: people evolve systems evolve we get mature technology evolves new things will come in. So those will be we'll consider okay this needs to be done let us put it in the backlog or take debt we'll address it when time comes but now whatever we have planned for this sprint we should have clarity on till one sprint level at least that is for the for the folks who are working on the floor for the leads like you and Manisha you should have slightly Rajashekar G: Good. Satyasri Prabhakar Mantripragada: forwardlooking uh features on now that the independent pieces are done what can be built on top of it what can be built on top of it. So you would have that vision and then you lay them the next steps. Rajashekar G: Right sir. Satyasri Prabhakar Mantripragada: So in in my mind this is this is clear for me. Okay. 00:59:26   Satyasri Prabhakar Mantripragada: This is how I'm executing it. This is how I wanted it to be executed. For me it is clear it is not clear. So where do you need my Rajashekar G: Okay. Satyasri Prabhakar Mantripragada: help functionality 100% that's okay but you should be able to demonst Yeshwanth Reddy Yerraguntla: H. Yeah. Satyasri Prabhakar Mantripragada: see that's where N was also correcting me don't think of it as a product that needs to work 100% for all the cases we are only building a platform to demonstrate that we can do this the platform is capable of doing all these it can accomplish these the these things. However, for each of the capabilities within that platform, if it needs to mature to a decent level, we need to have a customer buy in and the customer would ask how they want it. Then we'll buy time and customize to their needs and requirements. For now, we are not going to build a perfect system. That is fine. Name on record. 01:00:40   Yeshwanth Reddy Yerraguntla: Understood. I think both Gopalasar and you are telling the same thing that uh yeah we don't need there is no such thing as perfection at least in this stage. Whatever we do in its own way it's perfect is what both of you are trying to say in different uh vocabulary. Um in that case uh yeah we will try to uh we'll try to continue the sprint milestone one whatever we have committed to um that's what we'll promise by end of the sprint and milestone two end of the next uh I think milestone one is two sprints milestone two is uh four sprints from here now that there will be a change when genrex comes into picture and half the team is not involved. I want to understand now now I'm talking more in terms of the process now that like this Dinhana used to keep on telling me it never entered in my head at that time when I joined but now I'm understanding that it's all about the process now where in the process should we understand that there is a change in the team squad this thing there is that now that half of the team is doing something else. 01:02:16   Yeshwanth Reddy Yerraguntla: Um you know the milestones are going to be pushed, the milestone delivery timelines are going to be pushed. I would want ideally to have all of this somewhere in the sense that if I was Naven sir I don't want to come to uh either Rajar or Prabhakar or Yeshuant and constantly ask them hey what is going on? I would want ideally a place somewhere on the internet where all I have to do is okay I just go there uh I see what is the progress in the most ideal scenario on a day-to-day basis uh in the most ideal day-to-day but in ideal scenario at least I know what I am anticipating in the next two weeks in the next four weeks in the next six weeks for at a project to project basis Genhrx key this is what I know my team will deliver uh four weeks down the line uh for enterprise brain this is what I know it'll be delivered but tomorrow when I log in oh because half of the team went there this thing uh this promise has come down in the document now there is a new addendum saying because of so and so things um the milestones have been revised to so and so timeline Okay, I want to create a structure with this kind of a communication where somewhere we can 01:03:44   Yeshwanth Reddy Yerraguntla: be transparent to everyone of each other. That way when Rajka just said right like Jira was P5 but Ashantana was insisting that we do Jira that is a failure on my part also because despite me being in every meeting I I to have that kind of miscommunication is um it it just shows that how much uh we can goof up uh no matter how careful we Satyasri Prabhakar Mantripragada: I think uh I would I would just wanted to not the word correct but uh lack of a better word I'm using correct as a word but the thing is that whatever we started you me have agreed on a conversation that we would go with Google drive chat and then Gmail followed by Salesforce Salesforce has come in between but Yeshwanth Reddy Yerraguntla: H. Yeah. Satyasri Prabhakar Mantripragada: otherwise Jira was in the plan so from from your standpoint you are in the right direction only but thing is that meanwhile because of other business priorities things have Yeshwanth Reddy Yerraguntla: I Satyasri Prabhakar Mantripragada: got changed to a extent that now sales team or somebody started saying that okay if you are integrating Jira who is your target customer because we 01:04:50   Yeshwanth Reddy Yerraguntla: don't Satyasri Prabhakar Mantripragada: are going behind the manufacturing or non-tech industries who are not going to use Jira at all so there is no point you spending time and energy in integrating Jira rather if you have an ERP like a project and project portfolio management ment system pick up that and then do that whatever you wanted to do extract information from that so that would be more a selling point over Jira integration so that's where we had to pause it we had to take a call and then we wanted to say okay let us pause it now let's not spend any time on this that is for a P level that is fine and we crossed that P stage now and we are into implementation Yeshwanth Reddy Yerraguntla: know. Satyasri Prabhakar Mantripragada: now so the the the core concept, right? We wanted to demonstrate that we can build that capability of building agents, building communication using a chat kind of an interface which can consolidate information across different different sources. That part was already done. Whatever we have done with Gmail, GR, GSPACE and Salesforce, 01:06:06   Yeshwanth Reddy Yerraguntla: even Salesforce finished. We Satyasri Prabhakar Mantripragada: we demonstrated it. Yeshwanth Reddy Yerraguntla: demonstrated. Satyasri Prabhakar Mantripragada: Now with that demonstration we are getting clients who are asking for more in an accelerated fashion. Yeshwanth Reddy Yerraguntla: Yeah. Satyasri Prabhakar Mantripragada: So that's where we wanted to first crack the architecture on how we wanted to build Yeshwanth Reddy Yerraguntla: All Satyasri Prabhakar Mantripragada: and demonstrate so that we have a stronger voice to say in front of the customer that boss I know what I'm doing. I know how it is going to shape out. Today it is like this but tomorrow this is the bigger picture that it's going to be. So towards that we are all we are all building those smaller blocks and then integrating them together to build that bigger uh chasm. Yeshwanth Reddy Yerraguntla: right. Satyasri Prabhakar Mantripragada: So coming back to your question uh so in this case as of now one challenge that I see especially with Dwami in other engineering teams also except for those coming from maybe is that we are not actually Yeshwanth Reddy Yerraguntla: It's a 01:07:04   Satyasri Prabhakar Mantripragada: leveraging Jira to its fullest potential. Yeshwanth Reddy Yerraguntla: great Satyasri Prabhakar Mantripragada: Now I'm cons consciously forcing the team in in using all that even Navin that's a goal by Naven also that we should monitor the velocity. So unless you capture everything on a regular basis and update your daily status in on on a regular basis you will not achieve what you wanted to Yeshwanth Reddy Yerraguntla: Should Satyasri Prabhakar Mantripragada: achieve. So the systems are there but unless we feed in the right data you cannot get the right insights out of it. So here in this morning I was asking Rajar I'm not sure if I Yeshwanth Reddy Yerraguntla: I introduce myself? Satyasri Prabhakar Mantripragada: asked Manisha but I was asking them to fill in the due dates at the task level so that it it should also roll up to the story level and further to epic level. So Rashik, if you go to the timeline tab on the board. Yeshwanth Reddy Yerraguntla: Pleasure to meet you. Satyasri Prabhakar Mantripragada: So this timeline uh basically should give us refresh. 01:08:13   Satyasri Prabhakar Mantripragada: I can't see H. Yeshwanth Reddy Yerraguntla: No, I can't. Okay. Satyasri Prabhakar Mantripragada: So this should give us a picture on where things are. Where do we stand now? When is it expected to close? That is a question that I need to answer to the clients in this case Naru and Bretharu I should be able to say okay this will be done by March and this Yeshwanth Reddy Yerraguntla: All right. Satyasri Prabhakar Mantripragada: will be done by April end by May we have everything ready at least but I am not able to get that picture because I don't have any timelines linked to it we may have story points linked to it Yeshwanth Reddy Yerraguntla: H Satyasri Prabhakar Mantripragada: for effect track effort tracking but it should also have the target dates without neither you nor me nobody else can tell what when is it when is the project going to end so so this is how we will do it uh H1 however the management Yeshwanth Reddy Yerraguntla: This is Satyasri Prabhakar Mantripragada: may not be interested exactly in from the squad which person is exactly working on 01:09:16   Yeshwanth Reddy Yerraguntla: I Satyasri Prabhakar Mantripragada: this that need to handle I should be up to date on who is working on what they are more interested in how much of effort is spent across months or weeks And what is the cost involved in developing such a product? Yeshwanth Reddy Yerraguntla: One of the reason I wanted to force us to do Jira was this only that we will increase the transparency of what is going on on the for um yeah and whatever I'm discussing I am personally getting a couple of ideas on should we take this up in enterprise brain or somewhere else but yeah to to half of the battle I'm seeing with enterprise brain is it's not battle in that sense but to force ourselves to follow the process which in itself is a progress in my opinion because uh we are not going anywhere. Even though we were able to give really good demos for manufacturing, Salesforce, all those things, right? Satyasri Prabhakar Mantripragada: Yeah. Yeshwanth Reddy Yerraguntla: It there were bursts of uh brilliance uh that Satyasri Prabhakar Mantripragada: Yeah. 01:10:31   Yeshwanth Reddy Yerraguntla: happened. Satyasri Prabhakar Mantripragada: On that note, I was having this conversation with one of the engineering leads few months ago um sometime early last year or something. Somebody was saying that u why are somebody someone is insisting on the process um because I was also insisting on process because of some other past project experience within Diwami. So I was always that okay go link this bug to this test case only follow this workflow in Jira there is accountability there is tracing there is that governance and so on stuff and that person asked me why are you so what is what is it that you need do you need deliverable or do you need us to follow process simple and so I also thought okay for a while then I thought of giving an answer then I what I answered is that you can you can deliver excellent results at times. Maybe at that time the luck favored you or things were in your favor. But for you to be consistently delivering the same quality or same excellence, you need to know what is the process that you are following. 01:11:46   Satyasri Prabhakar Mantripragada: It is the process that brings the consistency in deliverable. On top of it is it is your brain that is giving you the better results. Yeshwanth Reddy Yerraguntla: And then Satyasri Prabhakar Mantripragada: But the cons for consistency you need to know what are the steps that you have done in the past that led to success. Yeshwanth Reddy Yerraguntla: um Satyasri Prabhakar Mantripragada: You just have to repeat those and improve on those. Yeshwanth Reddy Yerraguntla: Right. Satyasri Prabhakar Mantripragada: So for for younger folk sites it is it is like why is this guy behind this process? Why is he asking that what is it that you are getting out of it? Anyways, I'm delivering but for everything there is a system. The system is yielding some results. That's why organizations are following it. You just have to help in understanding the fundamentals behind it. Yeshwanth Reddy Yerraguntla: True. I see my wife right she works in PayPal and every day in the scrum they see the same Jira board even stone can be important why these things are important Not that it's not that without this we'll be lost or anything but with this so many other things uh can be achieved that we 01:13:17   Satyasri Prabhakar Mantripragada: You can course correct yourself when you see that you follow a certain pattern. You don't know whether it is right or wrong. But at the end of it when you retrospect you see that okay probably this is the place where I could have done better. This is the place where I have done it good. I should continue this. I should stop Yeshwanth Reddy Yerraguntla: Yeah. Satyasri Prabhakar Mantripragada: this. Yeshwanth Reddy Yerraguntla: Compounding effect. Satyasri Prabhakar Mantripragada: Yeah. Now coming back to the original discussion point. Vara Kumar Jagarapu: mute if they Gopal Gottumukkala: Oh sorry. So what I was saying is in that aspect right somebody saying I delivered by the process Vara Kumar Jagarapu: speak. Gopal Gottumukkala: that is not delivery. So the problem of not knowing delivery definition makes those statements I delivered. So what is the delivery is application working delivery? Oh certainly not. That is just one part of delivery and the code and where I am putting the point that this code is of quality screen is working great screen I 01:14:28   Yeshwanth Reddy Yerraguntla: emergency. Gopal Gottumukkala: can make screen great with 100 glitches in the code but where is where I'm putting that uh theory that my code is clean and going back how much time I took what was planned what is delivered Yeshwanth Reddy Yerraguntla: Who's Gopal Gottumukkala: What is left? Where I where have I written what is left? Yeshwanth Reddy Yerraguntla: next? Gopal Gottumukkala: It is a simple fact. If any sprint is delivered with a tech debt of zero, then somebody is lying. That is development. You cannot complete a sprint without creating a tech debt. So where have I kept it? If I'm saying zero then I'm lying. So if I'm not saying zero then what is it? and I move on somebody else comes in where he has to pick where is a hint for him that what was done right there are n such Yeshwanth Reddy Yerraguntla: I did drink yesterday. Gopal Gottumukkala: elements makes a delivery as delivery if we don't want to look into that we don't want to understand then I can make any kind of statement so it is an immaturity and the delivery aspect that comes out as a statement that it is working so why I need to fill the time why I need to fill and so on and so just a classic example right and one more thing before I forget month you don't change the timelines of milestones okay that disrupts entire thing what you do 01:15:56   Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: is when see you anticipated 40 hours a week but a week becomes 20 then take off the 20 out of the milestone don't change the timeline of Yeshwanth Reddy Yerraguntla: I also Gopal Gottumukkala: milestone that that has a Yeshwanth Reddy Yerraguntla: have Gopal Gottumukkala: bigger impact than this cuz once you start moving the date Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: that needle always moves that's not how it should be. So once you decided a timeline that means when world cup has to end by March 6th it it 6th it has to end by March 6th. If because of rain two matches are not being conducted drop them give them Yeshwanth Reddy Yerraguntla: I appreciate Gopal Gottumukkala: 00 give them 1. you figure out another alternative but that on that Yeshwanth Reddy Yerraguntla: it. Gopal Gottumukkala: day you know Ricky Martin has to come and sing he has to come and sing that that is that's the point of milestone milestones are not something that can be changed inside that bucket if there is something is causing Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: something to fall off it can fall off with a justification where do you write the justification if a epic is moving write in the epic that this epic is moving What we are doing in Jira, 01:17:10   Yeshwanth Reddy Yerraguntla: H Gopal Gottumukkala: we are just changing the date but we are supposed to attach a comment at the same time that we are not doing when I'm changing the status in progress I am supposed to put a Yeshwanth Reddy Yerraguntla: All Gopal Gottumukkala: comment when I'm painting to something else I'm supposed to put a comment Yeshwanth Reddy Yerraguntla: right. Gopal Gottumukkala: without it is again still half yes it changed nobody knows why because there is no other place to capture that information That Yeshwanth Reddy Yerraguntla: Good. Gopal Gottumukkala: ticket is a place correct meakar if I'm misquing anything. Yeshwanth Reddy Yerraguntla: That Gopal Gottumukkala: So what guys are doing there is a summary there is a description copy pasted summary there's nothing new in the description and ticket got opened in progress went to QA bug raised bug fixed and then uh QA certified done zero comments no change in the description then Jira is useless we would have taken Simple task management where the history maintenance nothing is there. I added a task. I completed a task. Then what is the point of Jira then there is no point paying for such a tool. 01:18:27   Gopal Gottumukkala: I'm saying that there is only a summary description is also equivalent to summaries. Vara Kumar Jagarapu: I Yeshwanth Reddy Yerraguntla: Is it Gopal Gottumukkala: And what it is going to tell me is on so and so date status changed. Yeshwanth Reddy Yerraguntla: chicken? Vara Kumar Jagarapu: don't Gopal Gottumukkala: That's all it is telling me. Then that is not information. That doesn't help anyone. So in our task estimation Jira is also delivery that gives the life cycle and that it the history has to be built in Jira so that the next guy when comes when he looks at the functionality he knows what kind of bugs are raised and if we can attach that PR number to it then he knows what was fixed what failed nobody needs to teach him Yeshwanth Reddy Yerraguntla: How's it Gopal Gottumukkala: anything it is just few critical bugs he knows where the heat in Yeshwanth Reddy Yerraguntla: going? Gopal Gottumukkala: the code where it is light where it is cool that is important for a debugger otherwise what happens is I open VS code I start putting break points that's the worst starting point for us to put anybody into that situation in order to avoid that situation this has to be done it is so important so you call it process process or you call it it is like when you are going into jungle right if you observe if anybody is following a new 01:19:51   Gopal Gottumukkala: path they will touch it's a bad thing to do they put some mark on the tree on the visible space when he goes back and turns back he'll make sure that that mark is visible from distance if not he'll go and put it in a different place of the same tree which is visible and then he moves on he moves So for him survival is important. Nobody needs to teach him he'll do it. For us we are not probably m paying attention to the product maintainability. That is survival of the product. Right? Product working is not the product survival. Yeshwanth Reddy Yerraguntla: Yeah. Gopal Gottumukkala: Once it is sold it has to be maintained. If it cannot be maintained, you know, it is only one time fee we will get then it is dead immediately. Yeshwanth Reddy Yerraguntla: All right. Gopal Gottumukkala: After that you already lost data. It is something like you know I paid a payment but there is nowhere it is written. I lost the payment that's all. There's nothing much can be done. 01:20:57   Gopal Gottumukkala: So Jira filling is also as valuable as not having a Yeshwanth Reddy Yerraguntla: H.B. Gopal Gottumukkala: transaction footprint in the database. It is a data lost. Got it right? We are only looking at time. We are only looking at velocity. We are only looking at something. No, we are not only looking at that. But those are the starting points to push people to get in there. Once you get in there, the next step can be requested. But ideally that that is how it should be done. And our estimates are considering this. We are considering all this and saying that okay this much time we'll do this much. So this is this is an obligation that we already accepted by taking a task. It is not something that is getting added on top. Yeshwanth Reddy Yerraguntla: Good. Gopal Gottumukkala: It is included. If not being included in our minds we have to do that because the great job that we are trying to do we are inherently saying except me nobody can use it. 01:22:01   Gopal Gottumukkala: That is like pretty bad situation to be in. Yeshwanth Reddy Yerraguntla: But you have to Gopal Gottumukkala: Who I mean for your problem Jer will be a best fit. Yeshwanth Reddy Yerraguntla: start. Gopal Gottumukkala: If not Jer equivalent to Jira. See I I am not in love with Jira but I am in love with what Jira can do. There are many other tools can also do but that is most important for us because anyway Jira is there industry Yeshwanth Reddy Yerraguntla: Right. Gopal Gottumukkala: already uh proved the point. We can spend time and pick another tool but there is no point because that tool also Yeshwanth Reddy Yerraguntla: Yeah. Gopal Gottumukkala: we have to learn anyway. The only thing is every time I'm doing put your decision there. Don't think the decision how good English bad English nothing matters. The guy who is going to look into it has a necessity. He has to fix a bug. He has to understand what was done earlier. He will spend that time correct that English and then try to interpret. 01:23:18   Gopal Gottumukkala: So we don't need to worry about how clean, how good it is. We keep doing it for a month. We know what we need to write. Also most of us are stopping because I don't know how to write a comment. Yeshwanth Reddy Yerraguntla: got Gopal Gottumukkala: I don't know how to write a description. Doesn't matter. Start. Yeshwanth Reddy Yerraguntla: more than how where is also one issue. I might search for the right place to do it. two things can happen. I might figure out the right place. So I'm happy I would have spent like probably two minutes I figured out well and good. But with the Jira what I observed is generally I try to do something I spend four five minutes Rajar can also watch for me how much we struggle with Jira in the calls. We'll just continue our discussion without touching Jira. Gopal Gottumukkala: Maybe take Praakaru's help. There are only two things Sakra and finding out the item that is one critical path. 01:24:18   Gopal Gottumukkala: Once we find the item, there are only two parts. Either what you are going to write belongs to code or it belongs to Jira. Nowhere else. If you are trying to if you think it has to be written somewhere else that must have been a ticket itself that's the basic principle. So how it is done that's not Jas how it is done is in the Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: code that is inline document also if that that what we are trying to do or close is a is a story itself then that is Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: a read me document of that particular module again in the code what Jira should talk about is what was the necessity whether it is done or not and point it to where that how What Jira should go is a comment and so and so module read me is Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: updated and probably link because it is you can always put a link because that link will never die right we just make sure that the living branch link goes 01:25:24   Yeshwanth Reddy Yerraguntla: Correct. Gopal Gottumukkala: there not the feature branch and that's all what Jan needs to be written is what you want to communicate to the owner of the ticket whoever is creating not how it is done. How it is done belongs to the code. And if I'm supposed to write an operational manual of that particular module, say this configuration has to be changed. Some something needs to be taken from somebody else. That operation manual belongs to the project documentation. Then that itself is a task for this module. Yeshwanth Reddy Yerraguntla: Hi. Gopal Gottumukkala: In that story there must have been a subtask saying that once this task is done make sure this document is updated again because it's a documentation task what and how has to be in this in that particular document and any complexities that you faced you need help those kind of things come to the Jira it's it's only a simple principle either it belongs to the target or it belongs to Jira nowhere else the moment It comes to us our mind that it has to be written somewhere else then there must be a task for that. 01:26:37   Gopal Gottumukkala: If we keep these two simple principles usually description we ch don't change right when we change the description when we are evaluating creating the acceptance criteria before it is solidified we keep amending the description once we start working it is mostly belongs to the comments and comments are about history no valuable information in the sense if we changed the logic mentioned in the comment I amended the description because of Yeshwanth Reddy Yerraguntla: All Gopal Gottumukkala: this reason on this point but actually that the change is supposed to happen in the description. These are the very few rules Jira doesn't have too many places to write summary description comment that's all and we need to figure out summary never ever change. Yeah, obviously we make a spelling mistake we'll correct. But usually if summary is changing that means the philosophy of the ticket is changing. Then must have a change request. Yeshwanth Reddy Yerraguntla: right. Gopal Gottumukkala: Theoretically description you when you are evolving when you are implementing either it belongs to inline code documentation or it belongs to the description that Yeshwanth Reddy Yerraguntla: Right. 01:27:48   Gopal Gottumukkala: means there is something changed during evaluation figured out. Put a comment that you know I'm changing change it there. Don't mention the change in the comment because what is preserved at the end even if you ditch comments the summary and description tells the story that's all Jira has it's not much beyond that everything else is about n number of dynamic variables and what you want to store in those dynamic variables everything else is to reach to that point in a number of perspectives That's all it has. It's a pretty simple system just because of dynamic variables. It feels complex because you have start date, you have due date, you have actual start date, you have actual end date, time in hours, time in story points. There's so many. But hardly in all those dynamic variables, you might need max four, not more than that. So we have to agree on what are the variables that we are targeting. Where do we write documentation? Where do we write comment? How do we write? 01:28:59   Gopal Gottumukkala: How do we change description? When do we change description? That's all what we need to agree within that particular group of team. Praakar if I'm going off please keep correcting me and don't wait. So and what I said earlier is no matter which task management even if you want to build task management also tell me if you get any bigger spec than this I'll be surprised everything Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: that you are trying to add is to reach to a ticket only no matter what kind of functional features that you add call it dashboards call it reports boards, Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: you know, call it sprint boards, velocity boards, no matter what you call, ultimately you are trying to reach to one particular ticket. That's all. Either you are trying to reach to the ticket or you are trying to aggregate number of tickets. That's all you are trying to do. All other things are being added. It's basically, you know, in data science, right? You have measures and you have dimensions. 01:30:07   Gopal Gottumukkala: So measures are facts, right? The fact is summary and description. Everything else is a dimension that you're trying to reach. And of course, three four c critical fields that you pay attention to are the facts. Everything else is dimension. You are trying to keep on creating dimensions. We can make an attempt. Actually, it's not that difficult. We can make an attempt. We'll make mistakes. But Raakaru is there. He will tell us okay this is not the way then that is the way then we'll adopt. three four max deviations. I think we'll be Yeshwanth Reddy Yerraguntla: Got it. Gopal Gottumukkala: l Yeshwanth Reddy Yerraguntla: Got it. Wrong usage. Wrong board is better than no board. Now I'll update that sentence to wrong usage of Jira is better than not Gopal Gottumukkala: Yeah. Yeshwanth Reddy Yerraguntla: using Jira. Gopal Gottumukkala: If it is wrong then there are people to correct us. So we don't need to worry. Yeshwanth Reddy Yerraguntla: Yeah. 01:31:13   Gopal Gottumukkala: Yeah. But remember those fundamentals and uh yeah all of us guys do not ask to change the milestone date. Always bargain if whatever reason it is when there is some effort has to be taken off you have to take off the lightweight features that again goes to the planning. So Prabhakar knows this science very well there is certain kind of formuladriven uh sprint planning. What we do is must things will only promise of a certain percentage of sprint. you keep some uh um some percentage of uh uh effort hours if something goes wrong they can be taken off so that your milestone is constant. Yeshwanth Reddy Yerraguntla: Okay. Gopal Gottumukkala: So I with practice it will come you need to have a project context delivery context what is being promised by what time if you have that context that planning will come right yeah a little bit of here and there always happens but that over a period of time we'll correct ourselves Yeshwanth Reddy Yerraguntla: Yeah, we'll try milestone like whatever currently we are committing. Let's not change the date. 01:32:39   Yeshwanth Reddy Yerraguntla: Let's bring down the number of features what we'll commit on Gopal Gottumukkala: Yes. Yeah. Bargain. Yeshwanth Reddy Yerraguntla: tomorrow. Gopal Gottumukkala: Every time that you plan a milestone right there also you you smartly you have to add something that can be dropped off without lot of business value. That smartness we have to apply. Satyasri Prabhakar Mantripragada: Yeah. Gopal Gottumukkala: We should not call out. You should never say it is not important feature. Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: Unless and until somebody is pushing so much that what is the value? What is the value? You don't highlight them. and somebody is saying it is so valuable for them but it is easy to implement. Don't say that it is easy to implement. Yeah, Yeshwanth Reddy Yerraguntla: M. Gopal Gottumukkala: I will do it. Okay. So that saves your your fireback scenarios to some extent. So especially for customers right when they're asking when we know it is very easily feasible don't say that word. Let me get back to you. 01:33:36   Gopal Gottumukkala: But seems like it is possible. If you want to give them confidence, say that come back and then let somebody else handle. Yeshwanth Reddy Yerraguntla: H. Gopal Gottumukkala: Why I'm saying somebody else handle? You, me, we don't have that knack. Okay. Yeshwanth Reddy Yerraguntla: Yeah. Gopal Gottumukkala: Kind of binary kind of answers comes out of our mouths. So best thing is come back and then tell our guys that it is pretty simple. accept it. They will figure out how to how to convey to the customer because they have another 10 items might be in the pipeline where the heat is right. They have to balance the heat. So they will have the contest they will balance that heat if we don't expose what what is uh you know the true effort of a task. These are small things we will learn but at least being developers we have to have a habit of not committing on the table. Yeshwanth Reddy Yerraguntla: H Gopal Gottumukkala: Give the person who can deal with that kind of commitments. 01:34:40   Gopal Gottumukkala: Let them deal with Yeshwanth Reddy Yerraguntla: got it. Got Gopal Gottumukkala: it. Yeshwanth Reddy Yerraguntla: it. Clear or not? I think one one action item for uh us as a developer team is to show up more on Jira. have more data entry points in Jira that whether something is important not important that's not how it should be but whenever there is a change whenever there is a update in a ticket not only are we updating the status but also in the form of comments second Gopal Gottumukkala: Yeah. Yeshwanth Reddy Yerraguntla: action item I'm seeing is um this milestone business we'll try to reevaluate with the bra cursor even though we didn't commit Gopal Gottumukkala: Yeah. Yeshwanth Reddy Yerraguntla: to maybe we did commit but now we'll say for once for the last time we'll re-evaluate and then actually commit to it sir we can do that tomorrow probably in the Satyasri Prabhakar Mantripragada: Sure. Yeshwanth Reddy Yerraguntla: evening I Satyasri Prabhakar Mantripragada: Sure. Yeshwanth Reddy Yerraguntla: think whatever are the milestones just have a single document saying this This is what the this is the expected deliverable in the milestone without any date. 01:36:04   Yeshwanth Reddy Yerraguntla: Put the document in viasa. We can use the document to as the base for communication. Rajashekar G: Okay. Gopal Gottumukkala: put a J link there it's not going to change Yeshwanth Reddy Yerraguntla: Yeah. Rajashekar G: Yes. Gopal Gottumukkala: right link dashboard link won't change put link see Rajashekar G: Yeah. Gopal Gottumukkala: benchmark if prabakaru asking what happened to the task after Yeshwanth Reddy Yerraguntla: Everyone's posting Gopal Gottumukkala: the promised date right see we would have missed his question Rajashekar G: All Yeshwanth Reddy Yerraguntla: and while we're waiting Gopal Gottumukkala: is not why you missed he don't know whether Mr. Yeshwanth Reddy Yerraguntla: on Gopal Gottumukkala: If that question is coming that means we missed something. See how many such questions we can reduce by weak. Rajashekar G: right. Yeshwanth Reddy Yerraguntla: those intros. Gopal Gottumukkala: Okay. If we had written something we had updated he's the best guy on the floor who can understand Jira and then grab that information. If he's not getting that information reach out to him. What would help him to understand? If we we think we did but he's not getting it. 01:37:07   Gopal Gottumukkala: That's the right time to ask what and how. Yeshwanth Reddy Yerraguntla: Yeah. Gopal Gottumukkala: Believe me, three four max we know what to do after that. Yeshwanth Reddy Yerraguntla: Got Gopal Gottumukkala: Three four max such cross questionings we should be in pretty good Yeshwanth Reddy Yerraguntla: it. Gopal Gottumukkala: shape. Yeshwanth Reddy Yerraguntla: We'll try it. Nice to meet you. Rajashekar G: You got it, sir. Yeshwanth Reddy Yerraguntla: Get together. I'm from the Android team Gopal Gottumukkala: See when you are um planning right if you need my help I I can participate. See one of the one of the major misses that we do Yeshwanth Reddy Yerraguntla: Got Gopal Gottumukkala: is we'll make our tasks too too bulky. So that bulkiness cannot bring clarity. Rajashekar G: All Gopal Gottumukkala: Make it atomic. We think why I need to spend 1 hour to break it. Rajashekar G: right. Gopal Gottumukkala: But believe me that 1 hour saves at least 5 6 hours down the line. Make a habit of that breaking to atomic level then clarity automatically comes. Even if we do all these things if our tasks are bulky then that clarity won't come. Okay. Yeshwanth Reddy Yerraguntla: it. Rajashekar G: Yep. Yeshwanth Reddy Yerraguntla: Yeah. Uh, Louisa. Gopal Gottumukkala: Anything else? Yeshwanth Reddy Yerraguntla: drop off. Nothing Gopal Gottumukkala: Yeah. Yeshwanth Reddy Yerraguntla: else. Rajashekar G: training Yeshwanth Reddy Yerraguntla: Hello. Rajashekar G: sessions. Yeshwanth Reddy Yerraguntla: Sorry. Gopal Gottumukkala: N is Yeshwanth Reddy Yerraguntla: Welcome to training. Gopal Gottumukkala: controller. Yeshwanth Reddy Yerraguntla: Uh I have been on this Rajashekar G: I've shared it. It's 2025 training. Yeshwanth Reddy Yerraguntla: background. So, this Rajashekar G: Uh, drive. Yeshwanth Reddy Yerraguntla: person Gopal Gottumukkala: Okay. So, I think I know that drive link. Uh me and Pan had a fight on it to correct the file names. So, probably that's what it is. Thanks. And thanks Brah. Thanks guys. Thank you. Yeshwanth Reddy Yerraguntla: by Bye everyone. Good night. Transcription ended after 01:40:00 This editable transcript was computer generated and might contain errors. People can also change the text after it was created.