Shreya Govil

After TikTok, this guy from IIT Madras wants to buy Google Chrome Aravind Srinivas, CEO of Perplexity AI, just made the tech industry's most audacious move: a $34.5 BILLION all-cash offer to acquire Google Chrome. Let that sink in. His company is valued at $18 billion. He's offering to buy Chrome for nearly double that. Here's why this matters more than you think: Perfect Timing: The DOJ is forcing Google to sell Chrome as part of antitrust remedies. This isn't just a random bid it's strategic. Chrome = Digital Real Estate: With 3.45 billion users and 65% market share, Chrome isn't just a browser. It's the gateway to the internet and Google's $300B ad empire. AI Browser Wars: Perplexity has already built "Comet," an AI-powered browser that can book flights, manage emails, and browse autonomously. Imagine that tech powering the world's most popular browser. This is just wow Per reports, multiple funds have offered to finance the bid with serious interest, even if the odds are long The Big Picture: This could be the AT&T breakup of our generation. If successful, it reshapes how we access information online and breaks Big Tech's stranglehold on digital distribution. The skeptics say it's impossible. Google will fight tooth and nail. Other tech giants will counter-bid. But here's what is mindblowing - Sometimes the biggest disruptions come from the most unexpected places. A 29-year-old from Chennai already built a billion-dollar AI company that threatens Google Search. Now he wants to own the front door to the internet. Whether he wins or not, Aravind just proved that in the age of AI, no monopoly is too big to challenge.

5,615

273 Comments

Stripe pays around 40-50LPA for SDE-1 and 80-90LPA for SDE-2 role🤯🤯🤯 This is the classic question they generally ask How to prevent double payment if API call fails? There are 3 possible scenarios: - Client sends request, but it never reaches the server – safe to retry. - Client sends request, server starts processing but fails midway – unsafe to retry. - Server successfully processes the request but fails to send response (network issue) – again, unsafe to retry. ⚠️ Retrying in cases 2 or 3 could lead to duplicate payments or multiple orders. So how do you handle this? 🛡️ Idempotency Keys to the rescue! 1. Client first requests a random idempotency key from the server. 2. Every time the client makes an API call, it includes this key. Server checks: 1. If the key exists in Redis (or similar in-memory DB), return the saved response. 2. If not, process the request and store the key + response. These keys have TTL (time-to-live) based on the use case. But what if Redis goes down? 😂 Let me know in comments… There are 3 more issues dekhte hai bta paoge ya nhi comments mein… #stripe #payments

3,559

229 Comments

Before your next SDE interview, make sure you’ve nailed these System Design concepts. (With resources) System Design interviews can be tricky and i got rejected in straight 5 companies in the start as i had weak fundamental knowledge 🥲 So, learn these before your next interview: 𝟭. 𝗙𝘂𝗻𝗱𝗮𝗺𝗲𝗻𝘁𝗮𝗹𝘀 (𝗖𝗼𝗿𝗲 𝗖𝗼𝗻𝗰𝗲𝗽𝘁𝘀)   • How the web works (DNS, TCP/IP, TLS, HTTP/HTTPS)   • Client-Server architecture   • REST vs gRPC   • Synchronous vs Asynchronous communication   • Stateless vs Stateful services 𝟮. 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗕𝗹𝗼𝗰𝗸𝘀   • Load Balancers (types, algorithms)   • Proxies (forward, reverse, CDN)   • API Gateway (rate limiting, auth, request routing)   • Web Servers (Nginx, Apache) 𝟯. 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲𝘀   • SQL vs NoSQL (When to use what)   • Indexes & Query Optimization   • Normalization/Denormalization   • Replication, Sharding, Partitioning   • CAP Theorem & PACELC   • Transactions, Isolation Levels   • Data modeling for large-scale systems 𝟰. 𝗖𝗮𝗰𝗵𝗶𝗻𝗴   • Why caching matters (reduce latency/load)   • Tools: Redis, Memcached   • Cache strategies: LRU, LFU, TTL, write-through, write-behind   • Cache invalidation strategies 𝟱. 𝗠𝗲𝘀𝘀𝗮𝗴𝗶𝗻𝗴 & 𝗤𝘂𝗲𝘂𝗲𝗶𝗻𝗴 𝗦𝘆𝘀𝘁𝗲𝗺𝘀   • Message Queues: Kafka, RabbitMQ, SQS   • Event-driven architecture   • Pub/Sub model 𝟲. 𝗦𝘆𝘀𝘁𝗲𝗺 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗼𝗻   • Service Discovery   • API Rate Limiting   • Circuit Breaker / Retry Logic   • Idempotency 𝟳. 𝗦𝗰𝗮𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗮𝗻𝗱 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲   • Horizontal vs Vertical scaling   • CDNs and Edge caching   • Database bottlenecks and read replicas   • Data consistency vs availability 𝟴. 𝗦𝗲𝗰𝘂𝗿𝗶𝘁𝘆 𝗕𝗮𝘀𝗶𝗰𝘀   • Authentication & Authorization (OAuth2, JWT, API Keys)   • HTTPS, TLS   • Data encryption in transit & at rest   • DDOS protection, throttling, rate limiting 𝟵. 𝗛𝗶𝗴𝗵 𝗔𝘃𝗮𝗶𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆 & 𝗙𝗮𝘂𝗹𝘁 𝗧𝗼𝗹𝗲𝗿𝗮𝗻𝗰𝗲   • Replication   • Failover mechanisms   • Leader election (e.g., Raft)   • Health checks, monitoring, and alerts 𝟭𝟬. 𝗣𝗿𝗮𝗰𝘁𝗶𝗰𝗲 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 You should know how to design:   • URL Shortener (TinyURL)   • Rate Limiter   • File Storage (Google Drive)   • Notification System   • Messaging System (WhatsApp/Slack)   • Social media (Facebook/Twitter)   • Ride Booking System (Uber) 𝗥𝗲𝘀𝗼𝘂𝗿𝗰𝗲𝘀 𝘂𝘀𝗲𝗱 𝗱𝘂𝗿𝗶𝗻𝗴 𝗽𝗿𝗲𝗽 👇🏻 1. Concept and coding YT channel 2. Hello interview YT channel and website. 3. Awesome HLD git repo 4. Gaurav sen's YT channel 5. Alex Xu’s System design interview book. Let me know if this is helpful to you in comments 🙌🏻 Please like and share for helping others preparing, and follow Ankur Kesharwani more. Connect with me for any kind of interview help -> https://lnkd.in/grXPhqVc #sde #jobs #faang #systemdesign #sdeinterview

1,955

65 Comments

I have gotten 30 referrals during my job hunt, and in the past few days there have been numerous DMs regarding how I did it. So I have recently switched roles from senior product analyst to Product Manager, in a span of 3 months. For companies like Razorpay , Meesho , Sprinto I was able to secure interviews via refferals. This is a guide to how I did it, so that anyone of you looking for jobs can benefit and make some steps ahead in the journey. Sharing everything you need to know Refferals (Save and share🚀): Whom to ask(in order of priority because this helps with faster reverts): 1. Direct acquaintances 2. Your LinkedIn connections 3. People who have hiring label in their profile, just go to comapny profile on linkedIn-> people -> hiring 4. People who are active on LinkedIn, like just check their last months activity. I tried aiming atleast 10 people for one company. Sample referral Notes ( these are the exact ones I used) : 📍While sending Connection request: Hey [Name], How are you? 
I came across an opening at [company] for [z] role, [job id]. I graduated from [college] and have [x] years of experience in the Product team at [company]. I find myself a right fit for the role, can you refer me for the same? 📍If connection request is accepted ( you don’t want to delay this, be very proactive here): Hey [Name], thanks for accepting my request. Sharing you my resume below, would be really helpful if you could provide a referral. 📍If they have shared resume with hiring team and you want to follow up: Hey [Name], I was wondering if you could advise me on how I can track the progress of my application at [company]? Do you have a specific portal or should I follow up with someone directly? P.S: Once I realised that only asking for referral isn’t enough, it’s equally important to follow up and keep a track that you actually have received a referral also. 💡This trick definitely increased my conversions of getting referred 2-3x. Do try this out! #jobs #referrals #samplereferrals #jobmarket #PMjobs #interviews #templates #jobhunt

805

38 Comments

The secret behind the rising profitability of Indian #startups can often be traced back to a single line item: the convenience fee. In FY 25: - Zomato : Profit -> Rs 527 Cr, Platform fee collected -> Rs 650 Cr - MakeMyTrip: Profit -> $95.3 M, Platform fee collected -> $110.5 M - IRCTC's convenience fee contributes to only 21% of revenue but 74% of profits For years, the belief was that “Indians don’t pay for services.” But that narrative is changing. Consumers today are willing to pay for the ease, speed, and added comfort they receive; so much so that even Amazon has started charging an extra convenience fee on #ecommerce orders. While these are growth engines but deciding on the convenience fee is extremely tricky. Take Zepto or Swiggy, where convenience fees are often the subject of heated rants. Yet, very few complain about IRCTC’s fee, which is significantly higher and has been around since its inception. So, what makes users tolerate a convenience fee in one case and resent it in another? An interesting lens through which to view this is the Delta 4 theory, popularised by Kunal Shah from CRED. It’s a framework for evaluating whether an industry or startup is likely to thrive, and it applies remarkably well to the psychology behind convenience fees. Here’s how it works: imagine you’re rating your experience of availing a service on a scale of 1 to 10. Take railway ticket booking, for instance. The traditional way, standing in long queues at a physical counter, would hardly get a 2 or 3. But booking the same ticket online via IRCTC? Most of us would rate that an 8 or 9. That gap, 9 minus 2 , is the delta IRCTC adds. And because it’s greater than 4, the #business sustains, and users rarely complain about paying a convenience fee. Now, apply the same lens to other industries: - Flight bookings: going all the way to the airport to book at the counter is a hassle, so paying ₹150–200 extra online doesn’t hurt. - Food delivery or groceries: Blinkit, or food delivery platforms, often face more resistance because the alternative (walking a few meters to a nearby restaurant or store) isn’t all that painful. So, the higher the Delta, the less likely people are to complain about the convenience fee. Convenience fees don’t operate in isolation. Product managers at Quick Commerce and food delivery #apps are constantly A/B testing factors like: - Average order value (AOV): Higher-value orders can absorb a small fee more easily. - Purchase frequency: High-frequency purchases tend to make the fee less noticeable over time. As these #tech startups mature, convenience fees are becoming the difference between #growth and profitability. They represent a new social contract: customers willing to pay for efficiency, convenience, and saved time. #kneeledge

810

59 Comments

🚀 System Design Prep ends here 🚀 🚨 I have cleared most of the system design interview rounds with this guide When it comes to System Design Interviews, there’s a lot of material out there… But trust me, you don’t need to drown yourself in 100s of resources. 👉 This one PDF is all you need to build a rock-solid foundation. 👉 I personally used it and cleared most of my system design rounds with just this. 👉 It covers the fundamentals + real-world architectures in a clear, structured way. 📌 Inside you’ll find: ➡️ Basics of scalability (Load balancers, Caching, Databases) ➡️ Deep dives into core components (Queues, Consistent Hashing, CAP theorem) ➡️ End-to-end architectures (URL shortener, Instagram, WhatsApp, etc.) ➡️ Tradeoffs explained in simple language 💡 Remember: In interviews, it’s not about fancy terms — it’s about clarity of thought, structured approach, and knowing trade-offs. This PDF helps you nail exactly that. 🔥 If you’re starting your System Design prep, this could be your one-stop guide. #systemdesign #interviewprep #softwareengineering #growth #careerdevelopment #google #microsoft

272

22 Comments

Solve these 7 Groups of 28 Sliding Window Problems to crack & solve mostly any problem designed for SDE Interviews on this pattern. I wish I had this cheatsheet back when I was first learning DSA. 1. Fixed-Size Window Generic Pattern: Move a window of fixed length `k` across an array/string. For each window, maintain a running calculation sum, frequency, etc. and update as you slide. Example Problems: - Max Sum Subarray of Size K GFGhttps://lnkd.in/esbNK-hJ - Maximum Average Subarray I https://lnkd.in/e3C-xidr - Substrings of Size Three with Distinct Characters https://lnkd.in/efMUP6xC - Number of Sub-arrays of Size K and Average ≥ Threshold https://lnkd.in/eKgk_-3e  2. Fixed-Size Window Requires STL/Advanced Container Pattern: Same as above, but need extra structures like deque, queue, hashmap to track max/min, negative values, or frequencies within the window. Example Problems: - First Negative Integer in Every Window of Size K GFGhttps://lnkd.in/eqnuBn-i - Sliding Window Maximum https://lnkd.in/eppz4fh9 - Find All Anagrams in a String https://lnkd.in/eVp-9rNH  3. Variable Size Window Generic Pattern: Expand the window until a certain condition is satisfied, then shrink from the left as needed. Used to find subarrays/strings that meet a flexible constraint. Example Problems: - Minimum Size Subarray Sum https://lnkd.in/ej9VutWu - Subarray Product Less Than K https://lnkd.in/etP75Tsm - Length of Longest Substring Without Repeating Characters https://lnkd.in/erc-fdgn  4. “AT LEAST” vs “AT MOST” Sliding Window Pattern: When asked to count subarrays with “at least” or “at most” k certain elements, extend the window as long as condition holds, then count the number of valid subarrays in one go. Example Problems: - Count Subarrays Where Max Element Appears at Least K Times https://lnkd.in/eSjtNDN9 - Number of Substrings Containing All Three Characters https://lnkd.in/e6BCcTy5 - Count the Number of Good Subarrays https://lnkd.in/eFC2n7X5 5. Exact K Using: exactk = atMostk - atMostk-1 Pattern: To count subarrays with exactly K distinct elements, odd numbers, etc., count “at most K” minus “at most K-1”. Example Problems: - Binary Subarrays with Sum https://lnkd.in/epDmuP7P - Subarrays with K Different Integers https://lnkd.in/eQrZ8znb - Number of Subarrays with Exactly K Odd Numbers https://lnkd.in/eT6u6Cwc - Count Complete Subarrays in an Array https://lnkd.in/emkKAuDK 6. Special Sliding Window Patterns Pattern: Custom requirements like finding all unique substrings, max/min values within window, or “circular” windows. Example Problems: - Find Repeated DNA Sequences https://lnkd.in/em5ZVE-3 - K-Radius Subarray Averages https://lnkd.in/eZ-n8vRs - Alternating Groups II https://lnkd.in/etF43HJP Continued in Comments ↓

224

15 Comments

I cleared LLD interviews at top companies like Intuit, Uber, Zepto, Groupon, SenseHQ, and Meesho! How? By sticking to these 4 focused steps 🔥 ✅ 𝐌𝐚𝐬𝐭𝐞𝐫 𝐭𝐡𝐞 𝐅𝐮𝐧𝐝𝐚𝐦𝐞𝐧𝐭𝐚𝐥𝐬 → Solidify your basics in OOP and SOLID principles ✅ 𝐃𝐞𝐬𝐢𝐠𝐧 𝐏𝐚𝐭𝐭𝐞𝐫𝐧𝐬 𝐌𝐚𝐭𝐭𝐞𝐫 → Creational: Singleton, Factory, Builder → Structural: Adapter, Facade, Decorator → Behavioral: Strategy, Observer, Command ✅ 𝐔𝐌𝐋 𝐃𝐢𝐚𝐠𝐫𝐚𝐦𝐬 𝐟𝐨𝐫 𝐂𝐥𝐚𝐫𝐢𝐭𝐲 → Class Diagrams: Attributes, Methods, Interfaces → Other Key Diagrams: Sequence, Use Case, Activity ✅ 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐞 & 𝐄𝐱𝐞𝐜𝐮𝐭𝐢𝐨𝐧 → Solved 8-10 complete LLD problems, applying design patterns and principles → Simulated interview conditions with a 1-hour timer ⏱️ 💡 LLD can be a game-changer for backend roles—it’s not just about writing code but designing scalable, maintainable systems that interviewers love! If you’re serious about cracking LLD rounds, join my Live LLD MasterClass where I’ll break down real interview problems and share proven strategies. 📅 Date: 1st March 2025 🔗 Register Here: https://lnkd.in/dx3fCtQz #LLD #LowLevelDesign #BackendDevelopment #SystemDesign #InterviewPrep #Uber #Meesho #Intuit #Zepto #Groupon #SenseHQ

82

1 Comment

Lenskart’s 70,000 Cr IPO: Growth Story or Overvaluation Risk? 👓 Lenskart.com is finally going public, and the number that’s getting everyone’s attention is its 70,000 crore valuation with a P/E of 235. For comparison, NVIDIA, the global leader behind the AI boom, trades at a P/E of about 60. So naturally, the question is, how does an eyewear company in India deserve a higher valuation than one of the fastest-growing tech giants in the world? Lenskart plans to raise around 7,200 crore through this IPO, but only 2,100 crore will go to the company. The remaining 5,100 crore will go to existing investors such as SoftBank, TPG, and Premji Invest, who are selling their shares. So, when new investors buy shares, a big part of their money isn’t going into Lenskart’s growth, it’s going to old investors cashing out. Now, there’s no doubt that Lenskart has built a solid business. It’s India’s biggest eyewear brand, designing, manufacturing, and selling its own products. The company operates over 2,500 stores in India and has expanded to markets like the UAE, Singapore, Japan, and the US. It's a fully integrated model that helps maintain quality and margins. Financially, the company’s numbers look good on paper. In FY25, it reported a revenue of 6,652 crore, up from 5,427 crore the previous year. It also turned a profit of 297 crore after a small loss of 10 crore in FY24. However, about 167 crore of this profit came from one-time income, meaning the actual business profit is lower. A few months ago, Peyush Bansal bought 4.27 crore shares from existing investors at an average price of 52 per share. Now, the IPO is valuing those shares at around 510, almost 10 times higher. While it’s common for founders to buy shares before an IPO, the large gap has raised questions about fairness for retail investors. Peyush Bansal has said the valuation is based on future potential, not current profits, and that might be true. India has nearly 500 million people who need corrective eyewear, but only a small part actually owns one. The growth opportunity is huge. So yes, it’s a well-built business with a long runway ahead. But paying 235x earnings still feels like a stretch. To me, the real question is : Is Lenskart’s IPO a sign of how far Indian startups have come? Or are we looking at another overvalued story, like what happened with Paytm? What’s your take? Would you invest in Lenskart at this price? 👇 LinkedIn LinkedIn for Marketing LinkedIn News India LinkedIn Guide to Networking LinkedIn News

62

If you’re preparing for system design interviews, read this BEFORE you waste 3 months. Here are the core topics you actually need to crack System Design interviews and understand real distributed systems: After going through DDIA, Alex Xu, blogs, cheat sheets, and countless threads… I realized we need a more structured, practical, interview-aligned roadmap. So I created one. 📌 1. Client–Server Fundamentals Most candidates jump to microservices without understanding the basics. Learn: single-tier vs multi-tier, stateless vs stateful, sync vs async, thin vs thick clients. 📌 2. API Gateways & Communication REST vs GraphQL vs gRPC, rate limiting, routing, retries, BFF patterns. This is the backbone of almost every modern system. 📌 3. Load Balancing & Traffic Routing L4 vs L7, consistent hashing, global load balancing, health checks. Interviewers love this because it tests tradeoff thinking. 📌 4. Databases & Replication Models Leader–follower, multi-leader, quorum reads/writes, sharding. This isn’t just “CAP theorem” — it’s knowing how real systems maintain correctness. 📌 5. Distributed Consensus & Coordination Raft, Paxos, Zookeeper, distributed locks, vector clocks, CRDTs. This is where DDIA helps, but still doesn’t give interview-level clarity. 📌 6. Caching Strategies (Real-world) CDNs, client/server caching, write-through / write-back / write-around, invalidation. Almost every “scaling” problem starts with caching. 📌 7. Messaging & Event-Driven Systems Queues vs streams, Kafka vs SQS vs RabbitMQ, delivery guarantees, DLQs. Critical for designing reliable, asynchronous systems. 📌 8. Storage Systems & File Architecture Object storage, HDFS, Ceph, tiered storage, replication, dedupe, backup & DR. Often ignored but extremely important in senior rounds. 📌 9. Observability & Reliability Engineering Metrics, logs, traces, SLO/SLI/SLA design, chaos engineering. This separates juniors from engineers who understand production. 📌 10. End-to-end Real System Case Studies Twitter feed, WhatsApp messaging, Uber dispatch, Netflix CDN, Stripe payments. This is how everything connects. The gap is huge. System Design interviews require: ✔ high-level reasoning ✔ tradeoff decisions ✔ bottleneck identification ✔ production constraints ✔ real-world architecture patterns DDIA teaches the internals — not the engineering mindset. 🎥 I’m starting a full video series covering each topic I’ll be learning each concept in depth with an AI tutor, asking every question you’d want to ask, and breaking down the topic from first principles → interview-ready → real-world. If you want clarity, depth, and zero fluff — this series is for you. #leetcode  #interviewprep  #dsa  #techcareers  #patterns  #softwareengineering  #systemdesign  #distributedSystems  #backendengineering  #highleveldesign  #hld 

150

2 Comments