Sumit Saraswat

Yeah. Sure. So, like, I've started my career with MetLife, and I have worked over there approx 7 years 7.5 years as a business analyst or data analyst. In 2019, I have joined Adobe System as a senior business owner where, uh, I was a part of a customer experience domain where we need to keep the track of the performance of the agents who are resolving the customer, uh, queries, uh, via through call or chat. And in Adobe, I have created, uh, like, different different, uh, Power BI dashboards. So we, like, used to get the data from the different sources, like our SQL Server, Excel, and different different JSON and KPIs, and we, uh, load those, uh, BI, uh, datasets into our Power BI system and build our visualizations based on the requirement of the stakeholders. So most of the time, it was a agent performance dashboard where we, like, create KPIs, like product 20 percentage, online percentage, and their contact per handled per day, contact handled per hour kind of a KPS we built on the visualization tool. Apart from that, I've also created the AUX usage report dashboard. Uh, it was totally based on the, uh, Power BI. And, uh, most of the time, we used to get the queries from the different different managers, from the different different teams that, hey. We need this agents, uh, this data from so we need to fetch, uh, that data, uh, from the particular uh, servers. So we, uh, and create the reports based on the requirement. So that's it from Adobe. Like, in 2,023 last year, I have joined Oracle as a senior business analyst. So, uh, in here, my roles, like, include, um, maintaining the dashboards, Power BI dashboards, and the OAC dashboards. Apart from that, we got the request from the stakeholders, hey, that you need to prepare some reports for the particular audience. And the partner side. So we prepare those reports and give, uh, uh, also provide the insights. So, uh, in here, in Oracle, we get the request through Jira. And when they, uh, we, like, explore the Jira request, we observe that, hey. What all are the KPIs they need or what all, uh, requirements they have? So basis on the requirements, we provide the data. I guess that's it from my end.

Uh, so it totally depends upon the requirements. So which method we use, it could be direct query or import. So if we want to improve our efficiency of the dashboard or report, so we preferred, uh, import because, uh, the all the data saves on your data model. But in in other hand, indirect query, the all the data saves on the server side. So we always, uh, if there's no need of, uh, like, daily refresh of the data or the requirement is not a urgent refresh. So we prefer that we can use import query. Uh, so import data. Uh, so what it did, it create a cache memory in Power BI Desktop and store the data in our model. So the retrieving of the data is very fast as compared to the direct query.

Yeah. I have worked on the, like, restrict the data to role level security. So for that, uh, it depends on the user. So if user belongs to the different different region or different different country, we provide the RLS, uh, based on their criteria. Uh, we use DEXs. Uh, like, if a particular user belong to a JPEG region, we provide decks, like, region equal to JPEG. And based on the role, we, uh, restricted them, uh, to use the rest, uh, specific data. Uh, so yeah. Uh, so it the rule level security is basically 2 types. Uh, one is, like, static and dynamic. In static, we just provide a like, based on the their, uh, role, where they belong, what kind of a database team they belong to. And in dynamic, we just pull their, like, username, uh, based based on the user principal name or username decks. And with the help of we provide the dynamic, uh, rollover security to the user. Uh, dynamic rollover security help us to, like, uh, if there are, like, thousands of users, uh, using the our dashboard, so we restrict them based on the dynamic rollover security. And, uh, we create a if they are, like, thousands of users, we create access. It based on their email ID. We map with our, uh, uh, and map we map though that table into our data model. And based on that, we pass the username with the help of the text and provide us, uh, dynamic row level security.

Should a particular post testing. Application security. No. I don't have answer for that right now. We implement, like, rule level security or object level security, but, uh, didn't understand this question completely.

What are the key consideration you would recommend? So most of the time, if we are working on a complex text, uh, we make sure that it should be concise. Variable, uh, should be there, and we don't use the indexes which are taking most of the memory. So we if they are, like, dexes who are, like, taking memory of the cache, we, uh, check with the performance visualizer. And if and if taking too much memory, we try to find, uh, those kind of dexes who are, like, taking less memory compared to the others.

When slow running complex text, how do you determine if it should be optimized for rewritten? If our current debts, which we have create is created, uh, taking too much time to execute, and we, uh, check the performance of the particular decks, uh, if it is, uh, like, good or it is not good as taking too much time in a performance analyzer. So we try to replace that with with the better decks.

Review the SQL query snippet below. There's a potential performance issue with it. Customers dot contribute. Yes. Uh, so, like, in here, we are using a wild card character to search for a specific country. Uh, so what can we do? We, like, replace that with equal to Germany. So it will help us to, uh, find, uh, that data where the country customer dot country is Germany Or, uh, second thing, uh, if, uh, needed, we can, uh, as per the requirement, we only face those columns instead of star or instead of asterisk. We only need to require those column which are required in our report. So it could be product ID, customer ID, revenue, whatever it is. We can fetch that.

I don't have answer for that. Oh, let me try, I guess.

Reporting. So isn't that accurately reflect complex? I don't have answer for that as well.