Arun Kumar

From cracking SDE 1 interview to SDE 3 at Swiggy: My Journey and Interview experience🚀👨🏻‍💻 Ever since I started coding, Swiggy has been a dream company for me❤️. Landing the SDE 1 role in July 2021 was a huge win, and today, I'm proud to be an SDE 3 – a result of the amazing growth opportunities at Swiggy.🚀 The interview process was definitely intense, but also a valuable learning experience. Here's a breakdown to help you prepare: The Interview Rounds at Swiggy: - Online Coding Test (75 mins): This round tested my problem-solving skills with 3-4 medium-hard coding challenges. Brush up on HackerRank, codechef, codeforces, GFG, Leetcode etc to get comfortable with this format.🚀🚀 - Machine Coding Round (120 mins): Here, I faced a real-world problem! I had to design a system, write code, and fulfill specific requirements within the time limit. This round tests your ability to think on your feet and translate ideas to code effectively. Practice designing systems beforehand and simulating real-world scenarios.🚀🚀 - Problem Solving & Data Structures (60 mins): This round focused on my coding fundamentals with 2-3 medium-hard coding problems. A strong understanding of data structures and algorithms is crucial here.🚀🚀 - Hiring Manager Round (45 mins): This round was more conversational. We discussed my past experience, system design decisions, and behavioral questions. Be prepared to showcase your technical expertise and soft skills relevant to the role.🚀🚀 - HR Round: This final round involved standard HR questions to assess your cultural fit and background.🚀🚀 I also discovered a challenge to upskill to land job offers like these. In the challenge, if you complete 90% of a course within 90 days, you’ll get a 90% refund. 🚀🫡🤞 Try it out here: https://bit.ly/3U3pFZB Will be sharing more valuable interview tips and insights in my upcoming posts♥️🚀 #swiggy #interviewexperience #journey #preplaced #mentor #pranjal

156

5 Comments

Below are some of the companies which are currently hiring for SDE 1 and SDE Intern role. 1) Adobe (SDE 1) 2) Rupeek (SDE 1) 3) Curefit - house of cult (QA role) 4) Stable Money (SDE Intern) 5) The Trade Desk (Software Engineer Intern) All the apply links are shared on our telegram and 1:1 WhatsApp updates. In case you are actively looking for a jobs and internship, do checkout our telegram channel. Join: https://lnkd.in/dgY77xN3 𝗪𝗵𝗮𝘁𝘀𝗔𝗽𝗽, 𝟭:𝟭 𝘂𝗽𝗱𝗮𝘁𝗲: 1) You will not missed out any job & internship updates. 2) You will get update only for your batch which are relevant to you. 3) it's 1:1 so we are also considering your privacy. 𝗛𝗼𝘄 𝘁𝗼 𝗷𝗼𝗶𝗻: (𝟮 𝗺𝗮𝗻𝗱𝗮𝘁𝗼𝗿𝘆 𝘀𝘁𝗲𝗽𝘀) 1) React and comment on the post 2) Fill this form: https://bit.ly/4bLlCbv Done, now you will receive all the updates on your WhatsApp!! Also connect with me if you wish Aman Chowdhury 🇮🇳

200

4 Comments

Swiggy used RAG to build its in-house Text-to-SQL -- allowing all employees to make data-driven decisions faster. 📈 Here’s how they built the system end-to-end: Suppose a user types a question in Slack, like "What's the average delivery time in Delhi last week?" The query passes through AWS Lambda, where a function cleans and formats it before sending it to the LLM. The model then checks the Knowledge Base for metadata related to “delivery time”, which includes tables and columns relevant to delivery data. It identifies “average delivery time” as the metric and retrieves relevant historical SQL examples. Using Swiggy-specific metadata, it selects tables like Orders and columns like delivery_time. The model also uses few-shot learning by adding reference queries from similar past requests to guide the SQL generation. Finally the generated SQL query is run on Snowflake, and gets adjusted if errors occur. The final query output is sent back to the user on Slack. Users can rate accuracy, helping refine the model. Link to the full article is in the comments. #AI #LLMs #RAG #SQL

153

8 Comments

My Google Interview Experience for SWE II role 💙 ❤️ 💛 💙 💚 ❤️ :  ⏰ The Timeline: 1️) Recruiter email on 15th April 2024 2️) Recruiter call on 18th April 2024 3️) Asked for some prep time (yes, you can do that!) 4️) Mock Interview on 10 May 2024 5️) 3 Coding rounds 6️) 1 Googleyness round 7️) 2 Team match rounds for the same team Based on my profile, the recruiter skipped the phone screen round and moved me directly to the onsite interviews. These onsite rounds are non-eliminatory, meaning I'll receive the results only after completing all of them. 💻 The Onsite Interviews: 1️⃣ Coding round 1: A topological sort problem with an alternative solution using color coding Difficulty: Medium-Hard - I let the interviewer know about both the approaches and  coded the first one but forgot to handle some edge cases. - I then resolved the problem and the interviewer appeared satisfied. - The interviewer then gave me a follow-up for the same question, to which I explained the approach and coded it as well. 2️⃣ Coding round 2: A problem involving Square Root Decomposition. Difficult: Hard -This problem was the most difficult problem among all my rounds. -I solved the problem in 30 mins; I had gotten confused with respect to the time remaining and hence coded it super-fast. -The interviewer then went on to ask me 5 follow-ups for this question. -I was able to solve 4 of them and code them as well. I got stuck in the last follow-up. I was really nervous about the result of this round but later on I got to know it had better feedback than my first round!!. 3️⃣ Coding round 3: Another Graph problem. Difficulty: Hard -This could be solved using a modification of Dijsktra’s algorithm. -The interviewer really checked my thought process and approach in this round and I had only 12 mins to code it up. Overall, it went well and the interviewer seemed very happy with the explanation and code. 4️⃣ Googleyness round: -I was asked 1 behavioural question with multiple follow up questions and different scenarios in each of the follow up. -I had prepared the top 50 questions asked in Interviews for this round. 💃 Result: -My interview feedback was very positive. I had a team match round 4 days after my Googlyness Round.  -I had a Fit Call with my to be manager and I liked the team so I conveyed the same to my recruiter. -Post that I had another fit call with another member of the same team, since this was a new team and they wanted me to meet the Team Lead as well. Once this was done I got my offer letter from Google within 5 days! 🤓 My Experience: The Google interview problems are very unique and they really force one to think out of the box. The whole process was very enriching for me and I had the best experience with the recruiters, interviewers and interview coordinators as well. They were very accommodating and always available!

3,773

118 Comments

Which all metrics needs to be monitored for restful webservices? 1. Response Time: Measure the time taken to process incoming requests and generate responses. Monitor average response time, percentiles (e.g., 90th, 95th, 99th), and trends over time to identify performance issues and bottlenecks. 2. Throughput: Monitor the rate of incoming requests and processed responses. Measure request and response rates per second or minute to understand traffic patterns, peak loads, and capacity utilization. 3. Error Rate: Track the percentage of requests that result in errors or failures. Monitor HTTP status codes (e.g., 4xx, 5xx) and error responses to identify issues such as client errors, server errors, or network issues. 4. Availability: Monitor the uptime and availability of the web service. Track service availability over time and calculate uptime percentage to ensure that the service meets reliability requirements and SLAs. 5. Latency: Measure network latency between clients and the web service endpoints. Monitor round-trip latency, DNS lookup time, TCP connection time, and time to first byte (TTFB) to identify network-related issues and optimize performance. 6. Concurrency: Monitor the level of concurrency or concurrent connections to the web service. Track the number of active connections, connection pool usage, and queue length to ensure that the service can handle concurrent requests efficiently. 7. Cache Hit Ratio: If caching is implemented, monitor the cache hit ratio to assess the effectiveness of caching strategies. Track the percentage of requests served from the cache versus the origin server to optimize cache performance and reduce response times. 8. API Usage Metrics: Monitor usage patterns and trends of API endpoints. Track the frequency of API calls, resource usage (e.g., CPU, memory), and API usage quotas to understand usage patterns and plan for capacity scaling. 9. Security Metrics: Monitor security-related metrics such as authentication failures, unauthorized access attempts, and anomalous behavior. Track security events, audit logs, and compliance metrics to ensure data protection and regulatory compliance. 10. Dependencies: Monitor dependencies and third-party integrations used by the web service. Track availability, latency, and error rates of external services, databases, and APIs to identify dependencies that may impact service reliability or performance. CONCLUSION:- By monitoring above mentioned metrics, issues can be detected, performance can be optimized, and ensure the reliability and availability of their RESTful web services. Some common tools available for monitoring these restful api are available like Prometheus , Grafana and many other. #backendEnineering #apis #systemdesign

6

Hey Linkedin Fam!! Thanks for showing interest. I have received a lot of responses. But I see most of the responders were not able to access the form. I checked, and have updated the form link. Updated form link: https://lnkd.in/gyk9RBxh Please fill the form. So that i can analyse the responses and proceed. If you face any issue, please feel free to reach me in DM or send me a connection request. I am accepting them all. But I can take responses only through the form. Hope this helps! #Hiring #SDEJobs #PrincipalEngineer #DevOpsJobs #CloudInfrastructure #KubernetesJobs #AWSJobs #SRE #Terraform #CloudNativeJobs #GoLang #Python #InfrastructureJobs #Docker #ServiceMesh #Helm #OpenTelemetry #GitOps

1

1 Comment

Hello LinkedIn family, If you are in Bangalore and looking for a house on rent, beware of NoBroker.com managed properties. These are the properties in which there is a mediator between you and the owner who calls himself a property manager. Why you should avoid such properties? 1. Even though there is a property manager assigned but no one hold accountability of any cleaning, painting or pest control. How would you feel if you are moving in tomorrow and the property is not cleaned, painted at all? 2. In such situation, property manager will stall you by saying - I’ll speak to the owner about it. But then nothing concrete happens. 3. The property manager is there to safe guard owner’s interest which is to get rent on time and if tenant does not pay rent then penalise him with extra money but there is absolutely nothing for tenant if things go south. 4. Even if you request to talk to owner directly to discuss the issues, they propose that they will arrange a conference call which too never happens. 5. The property manager will chase you only until you pay the deposit. Once you do that their job is almost done. I have been the kind of tenant who pays rent on 1st of every month. After I moved out, I told my previous house owner to take his own time to return the deposit money. I even managed the painting and cleaning of his house after I moved out coz he does not live in Bangalore. But what do tenants like us get when they try NoBroker’s ecosystem? Nothing but getting stalled on each issue. At last the property manager said - “Oh sorry about that but it was not my responsibility to get the house cleaned or painted. Owner said he will do it. But then he didn’t pay to our maintenance team and hence the house didn’t get maintained”. I am not sure if the property manager is serious or mocking me. NoBroker.com might be a good platform and it might be solving a common problem which is paying heavy brokerage. But if the property is managed by NoBroker.com team, please avoid it. Cc: Amit Kumar Agarwal

15

12 Comments

Sharing the list of FREE resources I used to transition from TCS to Intuit 🚀 I have been asked many times about the resources I referred to for preparation. So, here it is.👇 𝐅𝐨𝐫 𝐃𝐒𝐀 👉 I used GeeksforGeeks to consistently solve problems everyday for 2-4 hours after work hours. 👉 Applying filters on the topic and top companies list helped. 👉 I did not have takeUforward during my time, but you have Raj Vikramaditya. So, you can refer to Striver’s SDE sheet for preparation. 𝐅𝐨𝐫 𝐒𝐲𝐬𝐭𝐞𝐦 𝐃𝐞𝐬𝐢𝐠𝐧 👉 System Design Interview book by Alex Xu 👉 Gaurav Sen’s playlist 𝐅𝐨𝐫 𝐉𝐚𝐯𝐚, 𝐒𝐩𝐫𝐢𝐧𝐠𝐁𝐨𝐨𝐭 👉 Wanna read? - javatpoint.com 👉 Binge watcher? - Navin Reddy’s Youtube channel - TELUSKO 𝐅𝐨𝐫 𝐜𝐨𝐫𝐞 𝐬𝐮𝐛𝐣𝐞𝐜𝐭𝐬 Sometimes, companies also ask questions on OOPS, SQL, OS and Networking. I had watched Ravindrababu Ravula sir’s videos during my college days and prepared notes. So, I used to revise them before every interview. And last but the most important thing - keep going. Stay consistent.. Lage raho, manzil mil hi jaegi.. 🙌 Do follow Akanksha Buchke for hirings, referrals, interview preparation content and please mention in the comments any other resource you found helpful. PS~ Join this group for all tech content (DSA, coding, System Design - interview preparation and resources) - https://lnkd.in/gUw9-G_u #interviewresources #interviewexperience #interviewpreparation #interviewtips #softwareengineer #softwaredeveloper #referral #intuit #jobswitch #networking #connections #reach #linkedin #linkedincommunity #akankshabuchke

300

47 Comments

Yesterday, I cracked the Stock Buy and Sell - Multiple Transactions Allowed problem. In today's POTD of GeeksforGeeks challenged me with Stock Buy and Sell - Max One Transaction Allowed, where I learned how to optimize profit within a single transaction constraint by finding the minimum price and maximum profit dynamically. Every day is a new opportunity to level up! 👍 #VITBHOPAL #VITLions #DrGViswanathanChallenge #AlgorithmDesign #ProblemSolving #DynamicProgramming #GeeksforGeeks #CodingJourney

9

Hi Google Tech Team, I have been using Google Pay for a long time but 6 months ago, I deleted my Google Pay Account and then recreated with the same mobile number & same email address. After that, I am unable to fetch my credit score from Google Pay Application. Like, I am getting an error 'Cannot fetch credit report with this Google Pay account. If you have different accounts with the same phone number, sign in to the account you used previously and try again', while trying to access my credit report. As per the error message, how can I login to the previous account if I have used the same mobile number and email address in the new Google Pay Account. Logically, if somehow the details of old account data is still available in your database even though I have permanently deleted it from my end then it should not allow me to recreate a new account using the same mobile number or email address. Please have a look into this issue and fix it if it's a bug. #google #googlepayservices

5

APM-2 Experience at Meesho ⭐️ This was one of the companies I interviewed for during the job hunt process. Where did I find the opening? On LinkedIn How did I get the interview? Through referrals Rounds? Round 1: Intro round with group PM This was an intense round involving discussions around my candidature fit for Meesho and the work expectations. Round 2: Live SQL problem solving This was with the PM in the company to test SQL knowledge. It was a live SQL problem-solving round on Hackerrank. While I don't remember the actual questions, these were the topics: 👉Joins 👉Subqueries 👉WITH statement and sorting I was able to solve 2 out of 3 questions hence couldn't make it to further rounds. Follow me for deep-diving into real PM interviews. I’ll be sharing a detailed SQL learning guide in the upcoming posts. #interviews #jobhunt #layoffs #meesho #productmanager #pmjobs #experience

170

5 Comments

2022 → Joined Atlassian as SDE after completing internship. 2024 → Promoted from SDE to SDE-2🚀 This was my career graph so far and I know, many of you are aspiring to receive similar promotions. So, let me tell you some important skills to go from SDE to SDE-2 🔻 📌 **Critical problem-solving ability** - You need to find solutions quickly and ensure they are effective, implementable and scalable. 📌 **Hands-on DSA and System Design knowledge** - As you step up your position, you will have to learn advanced DSA and become familiar with System Design patterns. Let me give you a heads-up - Developing these skills while maintaining your 9-5 won’t be easy. But with the right support and 1:1 guidance, you can definitely get your dream position. Reserve your FREE session to begin your promotion journey - https://lnkd.in/gehT9h7X These SDEs at Preplaced will help you with DSA and System Design, provide constructive feedback, share valuable tips and assist in maintaining consistency so that you can get assured success🚀 #InterviewPrep #PreplacedCollab #Collab #Preplaced

160

5 Comments

Flipkart is hiring (2024 grads) 𝐑𝐨𝐥𝐞𝐬: SDE 1, UI-1 and Data Engineer 1 𝐄𝐥𝐢𝐠𝐢𝐛𝐢𝐥𝐢𝐭𝐲: 2024 engg graduates (CSE, IT, ECE & EE streams only) 𝐂𝐓𝐂: 32.57 LPA (18 lakhs base, 10% variable, ESOPs worth 5 Lakh, Joining Bonus- 3 lakh and Retention bonus: 3 lakh + retirals & other benefits) 𝐀𝐩𝐩𝐥𝐲: https://lnkd.in/ddTgFPam 𝗪𝗵𝗮𝘁𝘀𝗔𝗽𝗽, 𝟭:𝟭 𝘂𝗽𝗱𝗮𝘁𝗲: 1) You will not missed out any job & internship updates. 2) You will get update only for your batch which are relevant to you. 3) it's 1:1 so we are also considering your privacy. 𝗛𝗼𝘄 𝘁𝗼 𝗷𝗼𝗶𝗻: (𝟮 𝗺𝗮𝗻𝗱𝗮𝘁𝗼𝗿𝘆 𝘀𝘁𝗲𝗽𝘀) 1) React and comment on the post 2) Fill this form: https://Inkd.in/dQMqkbtz Done, now you will receive all the updates on your WhatsApp!! Also connect with me if you wish Aman Chowdhury 🇮🇳

144

8 Comments

The episode 11 of #ThisIsTechTalks is live on our channel, we talked about learning #springboot in 2024 with none other than Mr. SpringBoot Josh Long Watch the episode to learn how to get started with our co-hosts Santosh Yadav and Sonu Kapoor Link in the first comment: #spring #java

35

1 Comment

Around 3 Years ago I started my career with Swiggy as an SDE. If I had to restart my preparation in 2024 I will do these 4 things differently 👇 . 1️⃣ I’ll focus majorly on the current tech stack that the market is looking for, for example if they’re looking for react developer, I’ll study the basics thoroughly that is javascript. 2️⃣ Start solving 2 problems everyday on DSA to have a good grasp of concepts, because almost every interview will have atleast one dsa round. 3️⃣ ⁠Would create a couple of projects on the latest technology so that I can put those on my resume. 4️⃣ ⁠Start actively posting on LinkedIn to have a good reach so that it will have a slight advantage over other candidates. I believe these steps will significantly enhance your preparation journey in 2024 Note, CONSISTENCY is key. Put efforts in the right direction and you’ll get to your dream company. Now if you’re someone who is planning for a switch and needs help with the preparation then seek guidance from the ones who have achieved it. Check Bosscoder Academy Check here: https://bit.ly/4cXGltO They can help you with → ✅ Structured programs for learning DSA, System Design, and Full Stack Development. ✅ Personal mentorship from industry veterans. ✅ Live classes and 24/7 doubt support. ✅ Real-life, relevant projects with Placement Assistance. #DSA #Preparation #Collab

747

13 Comments

How Amazon S3 Achieves 99.999999999% Durability !!! 😨 💯 This post outlines how Amazon S3 provides a mind-boggling level of durability. Do check this out . 👇 #AmazonS3Durability

9

What are the top-paying / best engineering companies in #Hyderabad? - Google - Amazon - Microsoft - Uber - Salesforce - Qualcomm - Apple - Deliveroo - JP morgan - Nvidia - Servicenow - D.E shaw - Arcesium - Oracle - Zeta - Compass - Paypal - Oyo - Oppo - Indeed - Techmojo - NCR More - Altair - Celigo - Splunk - Turvo - Fanatics - Rudder Share the company you are working in? The only way $ in Hyderabad is gonna improve when there are more unicorn 🦄 startups establish here

31

𝐄𝐱𝐜𝐢𝐭𝐢𝐧𝐠 𝐧𝐞𝐰𝐬, 𝐞𝐯𝐞𝐫𝐲𝐨𝐧𝐞! 🔥  🔥 🔥 We have launched a brand new series to explore High level design in depth - entirely in Hindi. This series will include youtube videos(in hindi) along with notes and Medium Articles (in English). This series caters to both beginners and experienced professionals, providing a strong foundation in HLD concepts. 𝐒𝐭𝐚𝐲 𝐭𝐮𝐧𝐞𝐝! We'll be sharing valuable resources to strengthen your HLD knowledge. #HLD #Hindi #Design #systemDesign #Learning #Series #YouTube #Medium #Notes 𝐃𝐨 𝐜𝐡𝐞𝐜𝐤 𝐢𝐭 𝐨𝐮𝐭!

15

1 Comment

Having built a P0 service at the IPL scale during my first ever IPL at Jiocinema. One of the key learnings was to use caching at various levels. Having come from a B2B background where you can easily afford every request going to DB. In jiocinema it's just not possible in terms of infra and cost. You can't even afford every request to reach out to your origin server just because of the sheer scale of IPL. Those 200K RPS and 100K TPM transactions look cool but the real engineering lies in how you can design your application to minimize no of calls to your sever and database. The solution looks easy just use caching at every level. Caching at frontend. Cache at CDN. Cache at Application level. But when you are working at the IPL scale. How and what to cache becomes challenging. When you are operating at 20M concurrent users and 20% new users landing on your application every second . You cannot afford a call to the origin server on page/app reload specially during the peak . It will not only bloat up your servers but it will take down your cache/ cache servers and increase your cost drastically. If you start putting everything in cache your read/write time along with bills will shoot up drastically. Not to mention storing objects in the cache is a big No No. But at the same time, you must read from the cache most of the time> 90%. The second major challenge is how to scale your cache especially if you are reading and writing to it. Of course, when it comes to the IPL scale everything needs to be prescaled. We all have been a victim of cache not performing well on high throughput. Add write throughput in it and the problem becomes sheer complex. Having faced and solved these challenges and many more during my very first IPL at Jiocinema was quite a learning experience. Not to mention the pace at which we built and released this critical service. Edit 1 - If you are someone who hates putting an external cache in front of db. Go for database which provide a very efficient internal caching like ScyllaDb. #ipl #jiocinema #learnings #commerce #backend #engineering

73

3 Comments