I Solved 1583 Leetcode Questions Here's What I Learned

Summary

The video discusses common mistakes made while preparing for coding interviews using LeetCode. Key takeaways: understand fundamentals before solving problems; utilize LeetCode discussions to learn optimal solutions; stay organized with tools to track progress; practice mock interviews and contests to simulate real interview constraints. It's not the number of problems solved but the ability to tackle new problems that matters. Supplement LeetCode with real-world skills like open source contributions to stand out.

Takeaways

• 😵‍💫 Don't just copy code from Stack Overflow - understand the fundamentals and fix your skill issues
• 😊 Practice mock interviews and get comfortable with whiteboard coding - it's very different from Leetcode
• 📝 Use paper and talk out loud when solving problems - visualize and verbalize your thought process
• ✏️ Focus on understanding problems, not rushing to code a solution - think it through first
• ⏱ Use a timer to simulate real interview conditions when practicing
• 📈 Gradually increase problem difficulty as your skills improve - start easy and work up to hard
• 🏆 Doing contests helps prepare you for performing under time constraints
• 📊 Track your progress and keep notes on each problem for future review
• 🔍 Read solutions and discuss posts to improve your own approaches
• ❓ Measure readiness by your ability to consistently solve mediums in 25 minutes

Q & A

What mistakes does the speaker say he made when starting with LeetCode?

The speaker says his first mistake was not understanding fundamentals and jumping straight into solving problems. His second mistake was not utilizing the LeetCode discuss section to read optimal solutions. His third mistake was not doing LeetCode contests to practice in a timed environment.

What is the speaker's recommended easy/medium/hard problem distribution?

The speaker recommends solving 50% medium problems, 30% easy problems and 20% hard problems.

What is the recommended number of quality problems to solve according to the speaker?

The speaker recommends solving around 300 high quality problems.

What does the speaker say is key to getting the most out of LeetCode?

The speaker says staying organized is key for getting the most out of LeetCode. He recommends creating a list of attempted problems with status and notes.

What criteria does the speaker give for knowing if you are ready for coding interviews?

The speaker says if you can consistently solve LeetCode medium problems within 25 minutes and solve 3/4 problems in contests, you are likely ready for most coding interviews.

Does the speaker recommend using other platforms besides LeetCode?

No, the speaker says LeetCode is enough for practicing coding interview questions. He does not recommend rushing to get LeetCode Premium.

What is an alternative way to prepare for interviews besides LeetCode grinding?

The speaker does not mention this, but contributing to open source projects is an excellent way to prepare that demonstrates teamwork abilities.

Is the speaker's LeetCode advice applicable for experienced engineers?

The LeetCode practicing advice likely does not apply as strongly for experienced engineers. However, timed practice can still be useful.

What resources does the speaker recommend for learning data structures and algorithms?

The speaker recommends Geeks for Geeks, YouTube tutorials, and his own free algorithms course to learn fundamentals.

What is most important according to the speaker - number of problems solved or problem-solving abilities?

The speaker says the number of problems solved matters much less than how well you can solve new problems, which is why contests are very useful.

Outlines

00:00

😅 Struggling with coding interviews and improving through practice

The paragraph discusses the speaker's struggles with coding interviews and programming in college. He tried platforms like CodeForces and LeetCode but still felt unprepared. He recommends practicing by interviewing at companies you don't care about to get experience and build confidence.

05:03

😤 Don't use GeeksforGeeks, learn fundamentals first

The paragraph advises not to use GeeksforGeeks to learn. Instead, first understand basic data structures and algorithms before solving problems. Easy topics to start with are binary search trees, linked lists, stacks, and queues.

10:03

🤓 Be organized; use LeetCode Discuss section

The paragraph recommends staying organized with a list of attempted problems, their difficulties, status, and notes. It also advises reading the LeetCode Discuss section to understand different solutions, which helps get the most out of practicing.

15:06

⏱ Practice mock interviews with a timer

The paragraph suggests using a timer during practice sessions to simulate real interview constraints. It also recommends doing LeetCode contests for timed practice. Checking if you can solve medium problems in 25 minutes indicates interview preparedness.

20:07

👶 Don't just grind LeetCode; work on soft skills too

The paragraph critiques only preparing through LeetCode grinding. It advises working on an open source project to develop teamwork and communication skills, which are equally important in interviews.

Keywords

💡Programming Interviews

A major theme of the video is preparing for and succeeding in programming interviews at top technology companies. The video discusses mistakes made and lessons learned from the speaker's own experience with platforms like LeetCode and offers advice to the viewer on how to effectively practice and assess their readiness.

💡LeetCode

LeetCode is an online platform that provides programming problems and contests to help prepare for interviews. The speaker discusses both the benefits of LeetCode for practice as well as some common mistakes made in relying too heavily on the platform.

💡Coding Fundamentals

The speaker emphasizes the importance of deeply understanding core programming concepts and data structures before jumping into LeetCode problems. This includes things like basic algorithms, binary trees, graphs, etc. Strong fundamentals are key for effectively tackling problems.

💡Problem-Solving Approach

The video outlines a structured approach to solving LeetCode problems: 1) Understand the problem, 2) Write examples and map out a solution, 3) Code and test the solution, 4) Review other solutions. This kind of methodical process focuses on conceptual clarity first.

💡Assessing Progress

The speaker suggests metrics for evaluating progress towards interview readiness - e.g. being able to solve medium LeetCode problems under time constraints. Establishing goals and self-assessment helps avoid over-practicing problems unnecessarily.

💡Discussion Forum

The LeetCode discuss forums provide solutions and alternative approaches from others. Reviewing these after solving a problem yourself augments learning and exposes you to techniques you may not have considered.

💡Organizing Problems

Keeping an organized list of practiced problems, their difficulty, status (solved, need revision etc) and key learnings is recommended. This helps ensure you methodically progress across problem types and track areas needing improvement.

💡Contests

LeetCode contests simulate the time constraints and stress of actual interviews. Participating in these regularly provides exposure to that type of coding environment and pressure, which is useful preparation.

💡Premium Access

The paid LeetCode premium subscription provides access to more problems and company-specific questions. The speaker suggests the free tier is sufficient for interview prep initially. Premium helps target practice towards specific firms.

💡Soft Skills

While technical competence is critical, developing complementary soft skills are also important. Contributing to open source projects, for example, builds communication, collaboration and other interpersonal skills valuable for job roles.

Highlights

Practicing interviewing at places you don't care about is good practice and helps you discover new interests

Knowing fundamentals like binary search trees is important before diving into LeetCode

Solve around 300 high quality LeetCode problems before interviews

Start with easy LeetCode problems then gradually move to medium and hard

50% of solved problems should be medium, 30% easy, 20% hard

Understand the problem, visualize examples, write the solution out before coding

Read discuss solutions to learn optimal approaches after solving a problem

Stay organized with tools to track problem status and relevant links

LeetCode contests help practice under time constraints like interviews

Can consistently solve medium problems in 25 minutes indicates interview readiness

Focus on problem solving skills over number of problems solved

No need for other platforms apart from LeetCode initially

Contribute to open source to demonstrate team and communication skills

Pick an open source project that interests you and make several commits

Open source work helps develop professional skills beyond LeetCode

Transcripts

so apparently this guy solved eight 1583

leite code questions when I was in

college I was very struggling with

programming I tried code forces Cod and

even read dragging the coding interview

I never felt ready for interviews I knew

I needed to get better but I didn't know

how by the way a really great method to

get better is uh you're kind of worried

about interviews you feel like you're

not prepared for interviews go and

interview at places you don't care about

start applying to companies you would

never work at just to practice by

practicing and a you could find out

maybe you do actually want to work at

that company like that's the best part

is a you could discover something you're

really interested in doing and B you

could also you also just get that

practice you know there's no amount of

leak code you can ever do that's going

to make you feel comfortable standing up

in front of a whiteboard for the first

time there there is no amount of

pre-practice there's no amount of pre-at

there's no amount of anything to get to

that point you just got to stand in

front of a whiteboard to be comfortable

standing in front of a whiteboard CH in

2017 when I came across lead code I

instantly fell in love with its clean

and userfriendly interface and more

importantly the problems were actually

relevant to the real world interviews it

not only helped me improve my coding and

problem solving abilities but also

helped me read coding interviews at Big

tech companies like Amazon Google and

Microsoft in the past shambles right now

I'm literally in shambles I'm literally

in shambles right now okay they could

have yeah don't just don't even say it

don't even say Wy in shambles okay

um they took they took the end out of

fang he just did right there hurts me

few years I worked for more than 1500

lead quod problems mainly because I

didn't have anything else to do and I

feel really excited whenever

my okay it feels a little bad I mean uh

so what why did I do the ball I had

nothing else to do

so I just

built lead code

problems were accepted even for the easy

problems I thought more problems I solve

the bitter are my chances of cracking

interviews however that was Far From

Reality you don't need to solve these

many problems in this video I will talk

about three mistakes I made and the

lessons I learned so that you don't fall

into the same traps and increase by the

way you should do some of them so I

don't know what he's going to say I'm

curious what his three mistakes and what

you what you should do like once you've

once you're past like 1500 is so

aggressive but just knowing some of the

likely handshakes that you're going to

be asked like minimum subarray having

that one in your tuck is just so good

right because you know for a fact I've

been asked that two separate times two

separate times right it's crazy how many

times I get asked the same interview uh

questions you know what I mean your

chances of interviews my first mistake

not understanding fundamentals for some

of the topics I directly jump into

solving problems without understanding

how it works this is a big mistake

before you dve into lead code make sure

you have a basic understanding of

fundamental data structures and

algorithm for example know how to

implement basic binary s algorithms if

you don't how to do binary search off

the rip you're a loser okay I'm just

throwing that out there right now okay

Step It Up step it up it's simple you

should know it you should just know how

to do it

it is easy oh it's a tree you should

know how to do a binary search on a tree

if it's an array you should just know

how to do a binary search and you should

not get mixed up on where you do the

plus one it's just you just think

through it it's simple it's obvious it's

easy I'm sorry for being so mean to you

guys but sometimes you guys need that

Stern reprimanding okay wait hold on

whoa timeout time out time the the hell

out did I just say binary search and you

hit me with an easy array do sort fired

get him out get him

out get him out get the guy out just

Dr got it binary search easy boom ra on

sword dates likees linklist Stacks Q

okay you I I mean this guy's editing is

obviously really intense like he really

does a lot of editing but reting them

all is just giving me an actual

seure hmap binary PR articles grabs and

no algorithms like salting binary search

3D algorithm recurs and dynamic

programming DFS and BFS you don't need

to study any advanced data structures in

the beginning okay first off AVL trees

are pretty cool if you don't know AVL

trees you're also also loser okay avls

are freaking awesome okay by the way

this is not a proper AVL tree just so

you know no we don't no never mind this

is fine this is fine AVL tree it's is it

a fine AP L tree yeah I think it is in

the beginning for two reasons first they

will eat a lot of your time and second

they really ask in the interview you can

use websites like Geeks for geeks or

watch first off never use geek for geeks

that website is just

trash I hate that website I hate that

website like you look up anything AVL

tree okay and it shows up geek for geeks

watch this where is the geek for geeks

you hit that and what do you first get

boom you're going to get hit right away

with hey come on man get that b blocker

off so you say continue without support

crying face and then it hits you with

another thing and so then you exit that

and then sometimes it'll also show your

email address and it will also do this

and it will also do that and wait a

second rewind here rewind rewind I got

to rewind here first they will eat up a

lot of your time and second we're

rewinding and bf you don't need to study

any Advanced Data

12854 12854 yo don't don't take someone

else's diagram you got to at least

change it a little bit man okay you got

to at least change it just like a just

like a little bit just a little bit okay

just like a tiny just a tiny okay I mean

your point have been made your points

been made okay

but stres in the beginning for two

reasons first they will eat up a lot of

your time and second they really ask in

the interview you can use websites like

Geeks for geks or watch tutorials on

YouTube to learn

basic fundamentals you will find by the

way I do have that free algorithms

course I'll link it in the description

not a big deal NBD NBD NBD people it's

free it's N9 and a half hours long and

if you never sign up for front and

Masters I get paid zero cents so don't

worry I'm Shilling it for free for

freezy freezy easy peasy pumpkin

sey links to my favorite resources

there's no way he remembers put Des

problem lead code problems come in three

players easy medium and hard some folks

say you only need to solve 75 or 100 I

only do

hard because that's the state I'm

in problems but honestly it depends on

your coding experience but in general I

would say around 300 high quality

problems is a good number if you are a

new be like I was in the beginning you

need to solve more problems if you're

new to a topic start with easy problems

and gradually move towards medium and

hard problems it's like learning to walk

before you run here is my secret sauce

out of all the problems you solve make

it 50 one uh by the way every now and

then on Le code you'll get like an easy

problem or a medium problem that

definitely should have been a hard

problem like a toi a toi there's so many

little damn Edge conditions my goodness

it is just the worst I want that secret

sauce though 50% medium do you guys

agree with that honestly if I were you

and you were

me I

would do a good data structure and

algorithms course understand all the

fundamentals maybe do 50 leak code

problems easy and medium targeting array

style ones because honestly array ones

are going to be your hardest and they're

pretty common within it have a touch of

DP

experience and then go for the

interviews I think honestly failing

three interviews is no different than

doing a thousand leak code

questions because you learn what people

are asking DP stands for dynamic

programming you losers okay okay this

ain't hentai over here this is

programming get your in

order medium 30% easy and 20% hard if

you're solt on time in looking for a

curated ntop problems I highly recommend

solving lead quod top interview question

and 100 most like questions so how to go

about solving a problem first try to

understand the problem write it on a

paper visualize it in your head run

through a few input and output examples

think and write your Solution on paper

after you think your solution will work

run it on lead code and compare your

performance with others the last part

comparing your performance with others

I'd completely ignore that last part it

is utterly insane how they do

performance you literally can

just you run the same solution three

times and you can go from 100% to 0%

like instant

instantaneously um second off you should

always have paper come on dog look at

this I got paper and pencils right here

drawing stuff and thinking is a very

good way to go about stuff thirdly uh I

would actually treat it like an

interview if you really wanted the

practice get an effing whiteboard out

right do the problem and actually try to

solve it talk out loud do all of that

stuff put it on a whiteboard and see if

you can directly put it into leak

code and see if it actually um if it

actually does uh if it actually if it

actually even compiles if it even works

okay show us what's on the paper yeah

yeah yeah

yeah yeah here you go uh this one is um

my time my timings for when I do the oh

gosh I can't make it this is the timings

for when I do the algorithm course that

I have to have down in my head just in

case I forget the graph I want to use

for prims and csal here's a great graph

to be able to show some pretty cool

Parts especially with disjoint sets and

path compression uh planned it out

planned it out dog uh lastly here's

maximum flow Min cut here's the graph I

want to use so such that I will be able

to do an augmented path of purely only

using positives and then showing how you

can actually use the negative backtracks

to still continue to do augmenting paths

so I have two good two good graphs there

or uh two good ones and some notes you

know kind of like how you want to be

able to kind of teach these

things so anyways there you go nice

balls you're welcome Bud weird flex but

okay yeah it's the life I live I do look

at I do I do regularly study algorithms

algorithms follow Gooding prac believe

me naming aable ey during your interview

is not going to win you any points on

the contrary it can actually harm you if

you don't follow go to quing practices

during the interview don't too much time

on is that

true is that true as an interviewer as

someone who's done way too many

interviews do I really care how you

program on a

whiteboard nah for me the bigger thing

is just that you just talk through what

you're thinking I do like seeing you

solve the problem even if you solve it

wrong I still like seeing you attempting

to solve it I think that might be true

for a junior position I mean for a

junior position I just assume whatever

code they put into the into the code

base is going to be anyways and we

have to like help them right so why

would why would you ever assume that

someone's going to be a good like

especially a junior in fact I would be

more concerned for a senior if a senior

walks in and is just like making

addendums all over the place and it's

just wild then I go wow they really did

not pre-plan this at all in their head

they fully just went full stream brain

and just started ejaculating code and

they got it wrong right to me that'd be

more a single problem if you're un to

solve the problem within an hour then

look at the hints and the solutions go

to the discuss section and see like what

other solution people have posted this

brings me to the second mistake that I

made not utilizing lead code discuss

section in the initial days after

solving a problem I moved on to the next

one hoping that I have submitted an

optimal solution and it's good enough

for interviews but that was a huge

mistake really after you tackle a

problem read the official need quod

solution go to the discuss section and

try to understand top Ed Solutions you

can even contribute your own solution if

you think Your solution is unique stay

organized if that's I mean that's

actually pretty good advice I I could

see that seeing how someone else solved

it can always help you like there's

little things that you can do that just

make solving it especially on a

whiteboard much much easier uh honestly

it can make life really really it

there's a lot of things that are pretty

good why did he pick Java to trigger all

of us with light theme light themed Java

that was just he was there to this guy

chose violence he woke up and chose

violence you really want to get the most

out of lead code is staying organized

this key for that I

recommend uh junior control copy from

stack Overflow seniors over complicate

things you know I'm not going to lie to

you I hate this

take real talk if you're I know you

probably are just doing it as a joke but

I'm I'm leaning in okay we're leaning in

everybody get your get get your jackets

off okay we're going all in on this one

stay a little bit um is it the opposite

no if your job can be boiled down to

copying and pasting from stack Overflow

you are literally replaceable by an

AI

okay

stop stop your job is more than that

don't make I I I hate that joke because

it is so stupid I don't I haven't been

on stack Overflow in like two

years other than trying to ask dick

questions like I don't go to stack

Overflow pretty much ever to find things

and anytime I do I'm almost exclusively

disappointed by whatever they're showing

you come on people the only time stack

Overflow actually works out is because

you don't know the API of JavaScript and

you look up some simple question how do

I see if one string starts with a a

sequence of characters and it goes off

and goes stack Overflow how do I tell if

one and it just works okay yeah Prime

uses geek for geeks obviously I use geek

for geeks um that was please if you're

watching this in the future and you

didn't have the context that's a lie but

real talk you just simply like you

actually probably install left pad from

stack Overflow like that's the thing you

got to understand is that you use stack

Overflow because you don't know the

apis and if you keep looking up the same

stuff over and over again that's a u

problem fix the U problem okay just fix

it stack Overflow where pedants go for

validation their pedantry knows no

bounds skill issue you literally got to

fix the skill issue you are living a

life of skill issue fix

that loving a system When I Was

preparing for interviews I utiliz nosan

to create a list of problems with their

links difficulty levels status and notes

if I couldn't solve a problem in one

attempt I would change the status to

revise so that I can come back to it

later I ALS that's pretty good I mean I

think there's a lot of wisdom in this

that you should really try to keep track

of your progress but I would say that

the one downside to this is that uh you

are also keeping track of everything

you're doing which is a lot of busy work

right there is a lot of effort you have

to put into this um would I use obsidian

nah would I use notion no I kind of know

what I'm good at and I kind of know what

I'm bad at uh prob this is just not

really for me also shocking is that

there's this incredible level of detail

right here and then doesn't even name

the section okay hits them with the

Untitled did the camera die

F I'm going to turn this on for a moment

it'll probably overheat pretty quickly

but we don't have very much left okay

all right let's go let's go link all the

useful articles from the lead Cod

discuss section on the Bas with my

person provided in the description which

I use during my own preparation feel

free to duplicate it and add your own

problems according to your needs so now

let me talk about the last mistake that

I made not doing lead code contest for

the first few years I ignore lead quod

contest and only focused on solving

problems whenever I felt like it but it

was a bad idea and after realizing it I

started doing contest regularly during

my preparation I strongly advise

participating in lead code contest and

also virtual contest if you happen to

miss any because it helps you practice

in a timed environment after you have

solved enough problem and gain

confidence you start timing yourself

during your practice sessions this will

help you maintain focus and prepare you

under time constraint which is how the

actual interview happens so honestly

that's that's not terrible advice that's

you can use a timer you don't have to do

contest but I think contests really help

you put it puts yourself in the zone you

know what I mean because when you put a

timer up you don't have to compete

against anything but when you're in a

competition it forces that same feeling

it's actually not it's it's truly not a

bad idea I I really appreciate that

perspective I think that that was a

really well done perspective absolutely

this guy's cooking this guy's cooking

harder than my camera I like this I do

prepare how can you tell if you're ready

for the interview if you can

consistently solve lead code medium

problems within 25 minutes and three out

of four problems in the lead code

contest you are in a good spot for most

coding interviews remember it's not

about how many problems you solve it's

about how good you solving any new

problem that's why lead Cod cont are

very useful because they help youate

where you are in your interview

preparation Journey so do you need any

other platform apart from lead code for

practicing questions in my experience

the answer is no and don't rust to get

lead code premium to start with 200 to

300 problems and then decide if you

really need it one good use of code

premium is when you want to solve

company wise problem but that is not

needed in the beginning I hope it was

useful you can find all the links Below

in the description if you have any

questions or you want me to make a list

of my favorite Le Cod problems let me

know in the comment section I wish you

good luck for your interviews bye-bye uh

I like that that was actually really

good that was a very good video Ashish I

appreciate that I'm hitting you with the

sub hitting you with the sub press that

thumbs up um actually really I really

liked what he had to say he had a lot of

really good little bits of wisdom that I

think you forget when you've been out of

the industry for so or when you've been

in the industry for so long you forget

that one side uh there's a couple things

I think he did really well the timing of

the problems I think is a really good is

a really good practice because you

really want that forced stressful

environment because when you stand in

front of a whiteboard you got a timer

and not only do you got a timer you got

somebody watching but the thing that he

did not do uh well which is emphasize

other skills that are useful and

interesting the thing is is when you

come in and you're literally just like a

finite automata of just leak code you're

not interesting okay there's 10,000

people just like that invest in an open

source that seems interesting to you try

to get three commits in something that's

really really good right go look at

neovim and try to get three four five

commits uh I always think of TJ which is

like TJ is now a core maintainer and

he's just done an amazing amazing job of

going from not knowing even what git is

to becoming a core maintainer of neovim

and now he's had a lot of success

finding jobs he's working a great job at

uh at source graph he's doing language

server stuff like he picked a technology

got good at it and it was leverag for

him to work somewhere else that's also a

really really amazing thing to do and

you should consider that too because

once you also do that you will start

working with teams you'll start working

in a professional code base you will

start working with other people that

code differently than you and you will

learn how to work work in a team you

will learn how to communicate better

you'll develop all the soft skills that

you need it's a really good approach to

life um and so I really think that that

is just like a huge thing that he did

not cover and I'd really uh want to you

know kind of push on that hold on one

second sorry we got a poopy

diaper okay I was informed that there's

poop in a diaper

and I it might be mine it might actually

be mine okay anyways this is very very

good hey the name so I like that this is

good advice right here okay a Jen