TechInterviewPro Flashcards

Question

What are the **types of things** discussed in the **10-15 minute** part of one of the **coding interview?**

Answer 1

What **projects** they've done? How **deep can you talk** about the **projects technically**? **Time space analysis** details of **your project**? **Alternative versions** of **your projects**? How **deeply they can go** into **each project**? What their **interest level is**?

Answer 2

the **first 5 minutes** and **last 5 minutes** **matter. A lot!** These are your **first & last impressions**. Everyone says they're "passionate about coding," but if you **can show that passion by explaining about a project you've done that will be ideal.**

Answer 3

**Whatever they want** - There's **no "indoctrination" of interviewers**, people **just ask whatever they want**. People are **free to ask whatever they want**, although many **people will tend to just ask some standard question** that **they've heard about before**. they develop **2-3 questions**, and just **repeatedly ask** those **over and over again**. **key** is to show **how you solve it** and **dig deeper into the problem**.

Answer 4

as **someone** who is **collaborative**, **eager to learn**, and **humble** **Key traits:** * Eager **to learn** * friendly and **collaborative** * **technically curious** * able to **learn from mistakes** * willing to **accept feedback** and **improve oneself** * **get-it-done** attitude * **results-oriented**

Answer 5

**type of role** you are **applying for** There are many **different types** of **engineers**, who work with **different technologies**, depending on **varying organization size**

Answer 6

**People who can hit the ground running**, already having some **experience in technologies** like **Javascript, web development, iOS, Android,** etc.,

Answer 7

**Larger tech** companies may **prefer that you learn** their **proprietary tech stacks**. Must **demonstrate** **willingness, eagerness** to **learn!**

Answer 8

1. A situation where there was a **challenging failure** and how you **recovered from that through learning (learning from failure).** 2. How you were able to **accept feedback and change behavior** based on that (**accept feedback**). 3. Technical curiousity, enthusiasm and drive to learn (**learning curiosity**). 4. How do you deal with people you can't agree with and how to build consensus. (**conflict management**) 5. A project you are proud of or some recent work (**technically interesting project**). 6. Why do you want to work at this company? Make it person to you, specific company mission. (**What is your motivation**) 7. Take on personal projects, show to work on initiative (**Initiative and leadership**) 8. How were you able to tackle a situation with no clear answer? How did you **tackle the ambiguity** and **bring direction to the scenario**? (**Tackling ambiguous situations**)

Answer 9

**Start,** Challenge **Next**, what you did as an individual (action) **Finally**, what is the result of the action (result)

Answer 10

**Preparation** well rested, stories **clear and concise**, coding interview

Answer 11

**Situation**, **task**, **action**, **result** Use this to prepare stories

Answer 12

It means give **real, concrete stories** people **can dig into**! give sufficient detail **don't handwave** this! Prepare **before hand** and **hit key points**! Conflict resolution, building consensus, taking feedback

Answer 13

The ability to **take feedback** To improve **because of feedback** Most people to **jerks, arrogant, don't admit to shortcoming**!

Answer 14

The ability to **recover from failure**. Many times you'll build something you got it wrong. Conflict, things didn't go your way. A system crash, how did you recover? How can you **learn from the failure** and **prevent it**?

Answer 15

Come up with a **few stories.** Will dig more. What were **your motives**.

Answer 16

The ability to **gain consensus.** Teams **disagreeing on things such as prioritization**. How did you **build consensus**? What was **the process for gaining consensus**? Getting **people together**. Leading **a discussion**. Document where **people could voice opinions**. Stearing people **toward a direction**.

Answer 17

That we have **good stories**, **good actions**, good results That match the company philosphy (culture).

Answer 18

**Eagerness to learn**. Junior engineer's **immediate priority** is to **learn quickly** vs. cross team leadership, etc. Pretend you care and be curious! Show your curiousity.

Answer 19

**Hiring managers** want to **know**! Go into an interview **collaboratively**. **Smile**, lean forward, **positive energy**. **Build rapport**. Feel like you **belong there**. You can **hangout there.** They can treat you as a **normal employee.** Get them to **like you,** treat you **like a friend**. Don't need people to teach you, look out for you.

Answer 20

With a **first impression.** Get them to **like you,** treat you **like a friend.** **Smile, lean forward, positive energy.**

Answer 21

The ability to **learn is essential** Can you learn from **mistakes, failures** Facebook, Netflix, Amazon, Google - **using proprietary tech** stack, **learn quickly**. Struggle **a little bit**, work through **together**, **accept feedback**.

Answer 22

A process or system, not the employee find ways to fix the process or system. learn from **failure**

Answer 23

Do something out of your own **curiosity, initiative, interest**! Junior and Senior engineers **must have this**! **Show initiative through**: **side projects**, open source contributions, **pet projects**, **new programming languages** learned

Answer 24

**Time management** Make sure **you are on track**, cut chit chat, **save time** give you **time to talk of projects**, **get through white boarding** questions

Answer 25

End on a **good note** **Ask a good final question** First and **Last impressions** matter! The **way you talk**, dress sends **first impression** **Good questions:** What **new technology** is the **company exploring**? What is the hot new technology? What are your thoughts on **design patterns**? What other recent developments?

Answer 26

**Confident** You are an **expert at your craft** You are expected to **have competency** **collaborative**, **fairly confident tone**

Answer 27

Abstract away the **edge cases** **Abstract away code** to **manage your time** How fast can you do it, **solving the problem**

Answer 28

A strong introduction (30 seconds about you) **Discuss**: Talk about recent work, **summarize who you are** Most people read resume first time in front of you! They didn't study about you prior to meeting you.

Answer 29

Bring your **own marker pen** and **copy of resume**! Don't **rely on them** for a **marker.**

Answer 30

Highlight your **strengths** Create projects related to **software engineering** to show you are transitioning into software engineering

Answer 31

12 week **intensive bootcamps**! HackReactor Google summer hackathons **Structure**, **proper tech stack**, forced to put together a **final project.** Lots of people from **bootcamps** that end up **doing nothing**. It depends on **instructor**, **cohort**, **it's variable.** It's all about how **hard you work** and how **well you can pass interviews**.

Answer 32

get on the **right stack (proper technology)**! Setup some **pet projects**. Ruby on rails, python, Java, JQuery (**odd combination, not in demand**)

Answer 33

Bring your **unique** **background to a project** **Example:** If you are an **english major**, do a project on **natural language processing**. If you are a **BioMolecular engineer**, do a project on **DNA analysis**, etc.

Answer 34

Show **enthusiasm for coding** **Pet projects** are very important **Pet projects** convince interviewer that **you are serious** Validates why you want to switch

Answer 35

You can **exemplify your ability to perform multiple roles** Can handle **DB maintenence**, handle **project management**, **customer support emails**

Answer 36

Maybe start with **less salary** get some **project experience** **guide career** into **FANG company**

Answer 37

Study a **bit everyday** vs. cramming before the interview It's a good idea to **keep practicing** even **after getting a job** Keep the **skills sharp**! Churning out of companies can **help bump salary**

Answer 38

Where you are **weak** **Example**: Not knowing **python syntax** (want **real code**) make it runable cannot do **only pseudo-code** **Learn python**, **get syntax** correct **DS and Algorithms** - understand space/time complexity, how answer is gotten, understand every line of code, remember it, etc.

Answer 39

**Build a habit.** **Force self in first few weeks, then becomes habit** It's **hard at first,** then it **becomes fun** becomes like a **brain teaser**

Answer 40

Take a **few days off work** Grind hard for **12 hours a day** **Studying algorithms** **Solving problems** learning **about the company** **preparing stories**

Answer 41

Reach out to **recruiters** or **friends** in the **company** for **referrals**

Answer 42

**A good night sleep** **Get in a positive mood (affirmations)** **Get confidence up, perform better!** **Super important** You'll perform much better

Answer 43

Get a **good narrative** Had **technical background**, interested in **other field**, got into these **other thing**, realized **missed building/coding things** like **craft of coding**, have **pet projects to support this claim!**

Answer 44

Covers **construction of an entire system** (**end-to-end**) **senior engineer level** How do you design Twitter? What does **database** look like? What do **web services** look like? How do you **scale this**? What does r**ead/write API** look like? **Junior engineers** How to make it **multi-threaded**? What if you had **multiple cores**? How to **optimize** for **memory**? How to **optimize** for **time, space**?

Answer 45

Wholistic **system design (constructing a whole system)** design bitly design Twitter design Tinder How do you design a **distributed file system**? What **kind of database** (SQL vs. noSQL)? How is the **database setup**? How to do a **fetch, write**? What if you have need to **shard data** (many users)? How **do you shard**? What **are tradeoffs**? What is the key **you balance it on**? OS, **networks,** **database scaling** **How do you design web API?** Class with key values is there **privacy / security**? **error checking**? no bugs?

Answer 46

Give **you constraints** How do you **design around** constraints What is **in memory**? What is **in hard disk**?

Answer 47

How to do **calculations** (back of envelope) 500 ms request file 8 bytes meta data 16 bytes character string how big is a letter? (1 byte) string size? (256 bytes) integers? (4 bytes) boolean? (1 bit)

Answer 48

The **type of role** you're **applying for** **OS** engineer? **Web** scale-ability engineer? **android** engineer? **UI** engineer? **OS system** **design** interview is different than **web-scale-ability engineer** and **android engineer**

Answer 49

**Open ended** allows you to **show your strengths** show **your strengths** - API designs, database sharding How **broad can you go**? How deep can you go in **your specialization**?

Answer 50

Understanding the **goal for the system** **define:** use cases, constraints, assumptions What needs to be **prioritized**? (speed, reliability, availability) How many files will there be? How **many people** are **reading** from the system? How **many people** are **writing** to the system? What does the **load look like**? What is **response time** constraint? What about **data consistency**? Can have **stale cache data** or **up to date data** at all times? Do you need **value back right away** or **read comes later** on from a **stale cache**? **Who are the customers** using this? **How many people** are using the service?

Answer 51

**high level design** What are the **inputs/outputs of the system?** What **components do you need**? database, **load balancer**, **web server**, **services** (queue mechanism, notification services), overall **system layout**

Answer 52

Come up with the **components** know the **essential building blocks is key to A++ the system design interview** **load balancing, caching, CDNs, databases + indexing, database sharding, SQL vs. NoSQL, API design** break things down Is it reading from **cache or direct database**? How to keep cache updated? What happens if **system goes down**? How to recover the cache if system goes down? How to recover the **cache in system failures**?

Answer 53

A single **point of failure** in the **application** ## Footnote **Solution?** **Two, Three load balancers** **Website can resolve to different IP addresses, resolve to one of the load balancers, request can sent to server farm**

Answer 54

**Scale-ability** How do you **scale the system?** understand system and **how scaling options affect system** **Understand trade-offs (reliability, availability, security, maintainability)** **Ex.** load balancer introduces single point of failure. What happens to **reliability, security or maintainability of the system?** what if this **component is a bottleneck**? Will you **scale horizontally** or **vertically**? How do you deal with **running out of file space**?

Answer 55

One of the **key components** in any **web backend system** **Implementation?** AWS Elastic, NGIX **What?** **Balances the traffic load by distributing application traffic across multiple application servers** **How?** **Acts as a traffic cop** Efficiently distributing **incoming network traffic** across a group of **backend servers**, also known as a server farm or server pool * **Distributes** client requests or network load efficiently across **multiple servers** * Ensures **high availability** and reliability by **sending requests only to servers that are online** * Provides the flexibility to **add or subtract servers as demand dictates** **Why?** Improve **reliability** - if one server is down, can send traffic to another server Modern high‑traffic websites must **serve hundreds of thousands, if not millions**, of **concurrent requests from users or clients** and return the correct text, images, video, or application data, all in a fast and reliable manner. To cost‑effectively **scale to meet these high volumes**, modern computing **best practice generally requires adding more servers.** **Addition?** Can include **a cache**, send back **a request** before it pings an application server (faster response)

Answer 56

Creating **microservices** Instead of a **single server** to **handle all service requests** dedicate **one server per service type** Ex. chat server, website forum server, timeline feed service - application already **load balancing**, **hitting many machines, distributing traffic to microservice servers**

Answer 57

https://www.**nginx.com**/products/nginx/ best-in-class load‑balancing solutions used by **high‑traffic websites** such as **Dropbox, Netflix, and Zynga**. Purchase on **AWS marketplace**

Answer 58

* **Round Robin** – Requests are distributed across the group of servers sequentially. * **Least Connections** – A new request is sent to the server with the fewest current connections to clients. The relative computing capacity of each server is factored into determining which one has the least connections. * **IP Hash** – The IP address of the client is used to determine which server receives the request.

Answer 59

The **best way** to **reduce the load** on **your database** This is a **server** **Comes up in system design interviews all the time!** **Before you hit the database, you hit the cache server, if value is present client reads cache data without pinging database** **after hitting the database, cache server can store data requests (complex queries / joins) for future requests** **reduce latency** **Could be a distributed cache over multiple machines!**

Answer 60

**Write-through cache** - database is up to date, valuable data not lost, database write is slow due to cache layer First in first out - first item in is first out Least recently used Most recently used

Answer 61

When **user changes settings**, **remove cache** value **store meta deta** in cache **according to timestamp** **key hashed** to **users privacy settings** at a date or time

Answer 62

**very distributed, extremely scaleable** **Memcache** - free, open source (used by dynamic applications to eleviate database load) **Redis** is an open source (BSD licensed), in-memory data structure store, used as a database, cache and message broker. It supports data structures such as strings, hashes, lists, sets, sorted sets with range queries, bitmaps, hyperloglogs, geospatial indexes with radius queries and streams. Redis has built-in replication, Lua scripting, LRU eviction, transactions and different levels of on-disk persistence, and provides high availability via Redis Sentinel and automatic partitioning with Redis Cluster. **Apache Cassandra** is a free and open-source, distributed, wide column store, NoSQL database management system designed to handle large amounts of data across many commodity servers, providing high availability with no single point of failure

Answer 63

**Highly valueable** user data It can be **lost at any time** **used for caching** - redundent request gotten from source of truth, then **stored in cache** This memory is **not persistent**

Answer 64

**What?** **Provides distributed, very fast read and write into quickly** **Problem?** A **single database** under **high stress** introduces a **single point of failure, very fragile** **How?** **Can take a lot of load off the database** **store previous complex queries for future**

Answer 65

**cache and validation** mechanisms cache full (round robin, last used, custom string in entry) can have chron background job **perging useless cache data**

Answer 66

**Example**: You want to count number of visits to website each time you increment some value, ties up single database, very slow lots of writes, caching later, highly optimized as key value store

Answer 67

**cache** will **automatically propogate** changes **into the database** ## Footnote **write to database is slow because have to write through cache**

Answer 68

When we have **video**, **audio** or **image-heavy assets** that need to **reach users quickly**! **CDN's** can help scale.

Answer 69

**AWS Cloudfront** (**Amazon Web Service** CDN) **Softlayer** (IBM cloud)

Answer 70

**Web server** will **run faster**! Your server only **handles client requests!** Not serving **large image assets!** After heavy images and videos **offloaded server** performance increases!

Answer 71

A distributed **network of servers** that **host your static files** **What?** Images, videos, audio files HTML, CSS **Why?** **Fast access** to **large asset files** in **any part of the world** **As if locally stored in a region** **Who?** Most media-heavy website use CDNs

Answer 72

URL **resolves to an IP address** that is **geographically close** to **the user**! **Mechanism**? **pool** * store a cache of assets from your server in the region for future queries (easy to maintain) * slower first queries **push** * upload set of files that get pushed to CDNs * the first lookup is fast (pro) * have to manually upload files, must maintain repository (con)

Answer 73

Store data in **persistent disk** Has a **database server with an API to retrieve data** Query data, put data in, get data back **Database has an algorithm to retrieve data back** use **SQL queries** (delete, join tables, query range data)

Answer 74

Design a **schema** for a **database:** **relational** vs. **non-relational database** What are all the **tables you have**? What are **fields** in user table? What are **IDs / links** that **connect tables**?

Answer 75

**Algorithm for retrieving data quickly** **Indexes** - binary **search trees** based **on the field** **Every table should have unique ID**

Answer 76

A **binary search tree** based on **ID (primary key) or name** **Can do a binary search to identify in O (log n)** Can have **multiple indexes** (name, date) double index table Adds **space complexity** (binary search trees) Can **increase insert time** (have to update indexes)

Answer 77

It **adds space complexity** - have to build binary search trees for **each index** **Slows down inserts** - everytime you **insert** have to update **binary search tree (log n operation!)**

Answer 78

**ID to uniquely identify** it represent location w/ **longitude and latitude** represent date/time **using date (integer)** represent **userID** using **64-bit integer (scaleable)**

Answer 79

**don't** store **too heavy data** **move heavy data** (descriptions, large data) **out of database** insert into file system, CDN or separate table Ex. if use uploads photo of pet it's possible to store image in binary format in table but it boggs down table, makes database slower to query

Answer 80

**Joins** You want to **optimize joins** **using indexes** **(binary trees)** **identify joins in log n time!** **index on name of person** **index pet person by ID** **join on ID can find which pets, he owns fast!**

Answer 81

**Every table** should have an **ID field** that represents row **uniquely** (ID to **uniquely identify**) Move **large data** (picture files) into **separate file system** or separate table (not slow query times) **location** = longitude / latitude **date** = integer (timestamp) **userID** = 64 bit integer (scaleability)

Answer 82

Designing a **database schema** entries joining with **other entries** unique I**Ds on tables** **Indexies on fields** (fast queries)

Answer 83

**adding more machines** to the **system** (compute power) **doesn't work** with a **single database** **How does it affect fault tolerance, availability, reliability, efficiency, simplicity (keep in mind)?**

Answer 84

What **types of queries** am I going to be **making on this table**?

Answer 85

**Redundency and replication**! **Why?** **Usually one database** as source of truth cannot **throw a load balancer with many different databases**! Usually **just one table**

Answer 86

A very **popular way** to **scale database**! **mention i**n the **system design interview**! **Modes:** **Master-master replication** **master-slave replication**

Answer 87

Have **one master database** - write into **hundred+** **slave databases** - **read out only,** no writes, never write into, replicates master (slight latency) **When?** **Read-heavy application**, don't write often Ex. Youtube upload once in a while, **99% of people reading data**! Can have **scale slave databases**, scale this **way for reads**!

Answer 88

**Mutliple databases replicating eachother** acting as master soon as you **write into one database**, **replicates into other datbases** Can write into either database, can read from each can have muliple slaves **issues?** more complex to visualize slow writes to database (bottlenecks) trying to replicate into database, could be down have load performance issues **Alternative?** Use sharding over this method

Answer 89

**a slave** can **back up** the **master database**

Answer 90

A **write heavy** application Ex. Twitter people sending tweets every second **Why?** Master-slave solve read heavy applications **write-heavy applications** not solved w/ **master-master relationship, can't have hundreds of DBs replicating eachother!**

Answer 91

**Sharding (horitzontal partitioning)** - balance rows between databases (half ppl in database A, half in database B) **vertifcal partitioning** - separating tables out - move table to different machine, one table is on it's own machine, join according to userID with another table on a different machine

Answer 92

**database sharding** **Why?** the **sharding mechanisms** can be tricky (chosing where to right data, read data) **Alternatives?** **in-memory cache** like **memcache** can **scale out your reads**

Answer 93

**different databases** represent **portions of a single table** **distributing a single table** across **multiple databases** **every database contains a fraction of the data (table)**

Answer 94

**SQL pros** 1. **Data is structured** - can do range queries, has guarantees about data **SQL cons** 1. **NoSQL Cons** - 1. **Data isn't structured** - key-value dictionary data isn't structured, range queries are not possible, have to do queries based on key-values 2. **Not really consistent** - can take time for data to persist to disc, writes don't take right away, write delays, race conditions, data integrity lost. **NoSQL Pros** - 1. **Distributed** - distributed across many machines, scaleable across machines, can pump tons of writes into them 2. **Scaleable** - can handle intense data load 3. **MongoDB persists** to disk 4. **Graph database** - good for graphs

Answer 95

**Try to get** as **many offers** as **you can at once** ## Footnote **bring offers to others until you get same range for all**

Answer 96

When you **don't have leverage** **Competing offers** is **leverage**

Answer 97

**Leverage** **competing offers**, counteroffers from **current employer**

Answer 98

Being happy to receive one and NOT negotiating! **Just starting,** salary **will be lower.** Use leverage with **competing offers**

Answer 99

**optimize** for the **long term** **position yourself** well 10 years **in the future**

Answer 100

**Jumping jobs** **You have power** with an income the competing **company must match or beat**, potential **counter offers**

Answer 101

**Sign on** bonus **Move start date**

Answer 102

**exploding offers** not revealing offer unless you're **willing to accept it right away** **delays in interview process** at different companies (1-2 months Google vs. 1 week Startup)

Answer 103

Check the **Google Style** guides: https://google.github.io/styleguide/

Answer 104

**Salary** (yearly) **Sign on bonus** (paid first year, paid end of year) **Equity** (stocks) **Performance Bonus** (15% extra salary)