India version. Written for HLD rounds in India: estimates at UPI scale, and the RBI payment data and CERT-In logging and clock rules that shape designs built for Indian users. US version
System Design Interview Framework (India)
A timed HLD framework for Indian tech interviews, with UPI-scale estimates, RBI payment data storage and CERT-In log rules worked into practice designs.
The interactive tool loads here when JavaScript runs.
About this toolWhat is System Design Interview?
Check what your invite calls the round. HLD (high-level design) means services, data stores and scale; LLD (low-level design) means classes, interfaces and design patterns; machine coding means a small working program in a fixed time. This page is for the HLD round. Its four-step framework (requirements, high-level design, detailed design, scaling) fits a 45 to 60 minute slot, and the Practice Questions and Advanced Topics tabs give you prompts and building blocks to rehearse. For an LLD or machine coding round it covers only part of what you need.
Designing for Indian users changes the numbers and the constraints. A payments prompt here is a UPI-scale problem, and payment data falls under the Reserve Bank of India's 2018 storage rule: the entire payment data must sit in systems located only in India. The CERT-In directions of 28 April 2022 add three more design inputs for most companies: clocks synchronised to NIC or NPL time servers, cyber incidents reported within 6 hours, and logs kept for a rolling 180 days. The directions say within Indian jurisdiction, though CERT-In's FAQ (Q35) allows logs stored outside India if they can be produced to CERT-In within a reasonable time. For a payments or large consumer prompt, raise these during requirements, before you draw any boxes, because they decide where data lives and what you have to store.
The worked examples below turn those rules into numbers you can write on the whiteboard. Ticks on the framework and topics are saved in your browser only.
- Four-step framework with time boxes, steps and tips
- Six practice prompts with key components and focus areas
- About 40 core topics with short explanations, including CAP, QUIC, gossip and Merkle trees
- Checklists for framework steps and core concepts, saved in your browser
- Link to an Excalidraw-based diagram tool for architecture sketches
Updated
How to use System Design Interview
- Open the Framework tab and read the four steps with their time boxes before your first practice session.
- Pick a prompt from the Practice Questions tab and set a 45 minute timer.
- Work through the steps in order: requirements and estimates, then a high-level diagram, then one or two components in detail, then bottlenecks and scaling.
- Use the Advanced Topics tab to look up any building block you reached for but could not explain clearly, such as consistent hashing, Bloom filters or the saga pattern.
- Draw the final design in the Diagram Tool from the Whiteboarding tab, and tick the steps and Core Concepts you handled well.
Worked examples
Sizing a UPI-scale payment switch
Input
Premise given by the interviewer: 20 billion payments a month, 1 KB per ledger record, peak at 3 times average
Output
About 7,716 payments a second on average and 23,148 at peak; 20 TB of ledger data a month, 60 TB with three copies
20 billion divided by the 2,592,000 seconds in a 30-day month is 7,716 per second. The storage figure is the one that changes the design: 60 TB a month of replicated ledger data, all held in India, means partitioning by time and payer from day one and planning where cold data goes. Check NPCI's published monthly UPI statistics before quoting a real volume in an interview.
What the 180-day CERT-In log rule costs
Input
An API tier serving 5,000 requests a second on average, one 1 KB log line per request, logs kept 180 days
Output
432 GB of logs a day, about 77.8 TB over 180 days before compression (about 7.8 TB at 10:1)
5,000 x 86,400 seconds x 1 KB is 432 GB a day. Over the rolling 180 days that is 77.8 TB raw. Saying this out loud shows you know logging is a storage and cost decision, and that wherever the log store sits, you must be able to hand CERT-In any of those 180 days in reasonable time (its FAQ, Q35, allows storage outside India on that condition).
Processing abroad, storing in India
Input
A cross-region fraud model runs outside India on domestic card payments
Output
Offshore copies deleted and data brought back within one business day or 24 hours of processing, whichever is earlier
RBI's FAQ on payment data storage allows processing outside India but requires the data to be deleted abroad and brought back within that window. In a design this becomes a 24-hour time-to-live on the offshore store and a reconciliation job proving deletion, which is a good trade-off discussion for the scaling step.
Background and insight
System design rounds are open-ended on purpose. There is no single right answer, so the interviewer is scoring how you get to one: whether you pin down requirements before drawing boxes, whether your numbers are in the right order of magnitude, whether you can go deep on one component, and whether you name trade-offs instead of presenting one option as free. Seniority shows mostly in the last two.
Back-of-the-envelope numbers anchor everything. A day has 86,400 seconds, so 100 million requests a day is about 1,160 per second on average, and peak is commonly two to three times that. A 7-character Base62 short code gives 62⁷, about 3.5 trillion, combinations. A single SSD-backed database node can usually serve thousands of simple indexed reads per second, so 1,000 QPS rarely needs sharding, while 100,000 usually needs caching or partitioning.
The CAP theorem is often stated as "pick 2 of 3", which is misleading. Network partitions happen whether you choose them or not, so the real decision is what the system does during a partition: refuse some requests to stay consistent, or keep answering and risk stale reads. PACELC extends this with the latency versus consistency trade-off that applies even when the network is healthy, and it is a better lens for most interview discussions.
Common use cases
Preparing for a senior engineering loop
Practise one prompt every few days with the four-step framework until the timing feels natural.
Filling knowledge gaps
Use the Advanced Topics list as a checklist of building blocks you should be able to explain in two sentences.
Running mock interviews
As an interviewer, pick a prompt and use its focus areas as the probes for the detailed design step.
Reviewing real designs
Apply the same requirements, estimates and trade-offs structure when reviewing a design doc at work.
Tips and limitations
- Ask early whether the system handles payment data. If it does, say the RBI storage rule applies and keep the primary and disaster recovery copies in Indian regions; AWS, for example, lists two regions in India, Mumbai and Hyderabad.
- Put time synchronisation on the board for anything with ordering or audit needs. The CERT-In directions name NIC and NPL NTP servers, or servers traceable to them.
- Peak traffic in India clusters around salary days, sale events and festivals. Ask the interviewer for the peak multiplier instead of assuming the usual two to three times average.
- Name the 6-hour CERT-In reporting window when you discuss monitoring: it means alerting and on-call have to detect and triage an incident fast, not in a weekly review.
Sources and further reading
- Storage of Payment System Data (RBI/2017-18/153), Reserve Bank of India. The 6 April 2018 direction that the entire payment data is stored in systems only in India.
- FAQs on Storage of Payment System Data, Reserve Bank of India. Processing abroad allowed; data deleted and brought back within one business day or 24 hours.
- Directions under sub-section (6) of section 70B of the IT Act, 2000, CERT-In, Ministry of Electronics and Information Technology. NTP synchronisation, 6-hour incident reporting and 180-day log retention.
- FAQs on the cyber security directions of 28 April 2022 (May 2022), CERT-In. Q35: logs may be stored outside India if they can be produced to CERT-In in a reasonable time.
- AWS Regions, Amazon Web Services. Lists the Mumbai and Hyderabad regions in India.
Frequently asked questions
What is the difference between HLD and LLD rounds in Indian interviews?
HLD (high-level design) is about services, data stores, APIs, scale and trade-offs, which is what this framework covers. LLD (low-level design) is about classes, interfaces, relationships and design patterns for one component, such as a parking lot or an expense-splitting app. Machine coding asks you to build a small working program in a fixed time. Ask your recruiter which of these your loop includes.
Do I need to mention RBI data localisation in a payments system design?
Yes, if the system stores payment data. RBI's April 2018 circular on storage of payment system data requires system providers to store the entire data relating to payment systems only in India. Its FAQ allows processing abroad if the data is deleted there and brought back within one business day or 24 hours. Mentioning it during requirements shows you understand the constraint shapes region choice and disaster recovery.
How do the CERT-In directions affect a system design answer?
Three of them change the design: synchronise clocks to NIC or NPL NTP servers (or ones traceable to them), report listed cyber incidents to CERT-In within 6 hours of noticing them, and keep logs of all ICT systems for a rolling 180 days (the directions say within Indian jurisdiction; CERT-In's FAQ, Q35, allows storage abroad if the logs can be produced in reasonable time). That gives you a time source, an alerting requirement and a log retention budget to size.
How should I estimate scale for an Indian consumer app?
Start from the user numbers the interviewer gives you, then ask about peaks: sale days, cricket matches and festival evenings can be many times the average. Work out average requests per second (daily requests divided by 86,400), apply the peak multiplier they confirm, and size storage and logs separately, since the 180-day log rule can make logs larger than the main database.
Which practice prompts are worth adding for Indian product companies?
Besides the six on the Practice Questions tab, rehearse a UPI-style payment switch, a food or grocery delivery backend with live order tracking, a ticket booking system that survives a flash sale, and an OTP or SMS notification service. Each forces a trade-off this framework handles well: idempotency, geospatial lookups, contention on limited inventory, and third-party rate limits.
Is there an in-depth guide to System Design Interview?
Yes. “System Design Interviews: A Four-Step Method with Real Numbers” covers the background, how it works, common pitfalls and worked examples. Read the full guide.
Related Tools You May Like
Free, in-browser tools that pair well with this one.