DEV Community: Soham Malakar

Is the LeetCode Grind Overrated? My Unexpected Google Interview Success Story

Soham Malakar — Mon, 21 Apr 2025 16:41:47 +0000

Your friendly neighborhood code wrangler here has officially clocked in a full quarter at the Googleplex, and let me tell you, the rumors are (mostly) true - free snacks vanish at warp speed, and the internal jargon could rival a Vulcan dictionary (my apologies to the Star Trek purists). Beyond these quirky realities, however, lies a Google hiring process that, while often shrouded in mystery, possesses a surprisingly human element. The internet is filled with advice on how to crack this coveted interview, and a significant portion often revolves around the diligent practice of LeetCode problems. While I won't deny the value of understanding fundamental data structures and algorithms - skills that LeetCode undoubtedly helps develop - my own path to Google suggests that an over-reliance on this platform might lead aspiring candidates to neglect other equally important aspects of the hiring process. My journey involved a more balanced approach, focusing not just on algorithmic proficiency honed by coding platforms, but also on clear communication, system design thinking, and demonstrating that elusive "Googlyness." Whether you're dreaming of debugging distributed systems or just want to finally understand recursion, these insights, gleaned from my 2+ years of pre-Google 'experience' (mostly figuring out where the good debugging tools were), should arm you with more than just textbook knowledge.

Act I: The Interview Gauntlet

Scene from Silicon Valley (TV show) via HBO

Telephonic Screening: The HR screen wasn't just behavioral; it delved into the core principles of data structures and algorithms, probing my understanding of time complexity and trade-offs in a way that went beyond basic definitions. For the ML track, they focused on foundational concepts like bias-variance and the practical implications of different model evaluation metrics.
Technical Interviews: The DSA round wasn't just about solving problems; it was about demonstrating clear thinking under pressure and articulating the why behind my chosen approach. The high-level design interview felt less about knowing the 'right' answer and more about showcasing my ability to think holistically about scalability and system architecture, even when faced with ambiguous requirements. The ML rounds pushed beyond theoretical knowledge, challenging me with real-world scenarios and requiring me to justify my model choices and discuss potential challenges in deployment.
"Googlyness" Round: The 'Googlyness' round wasn't about reciting corporate values; it was about understanding how I'd navigate team conflicts, approach collaborative problem-solving, and demonstrate a genuine curiosity for learning and growth. For instance, when asked about a time I disagreed with a colleague, I focused not just on the disagreement but on the process of respectful communication and finding a mutually beneficial solution.

Act II: My Semi-Serious, Mostly Sarcastic Survival Guide

Scene from American Psycho (2000) directed by Mary Harron

The "I'll Prepare Forever" Fallacy: Break the cycle by setting a realistic application deadline for yourself. Treat it like a mini-project. Instead of endless prep, focus on targeted learning for a set period, then apply. The feedback you get from actual interviews is invaluable and often more insightful than another month of isolated studying.
Speak Your Brain, Don't Just Mumble Code: Think of the interviewer as a collaborator on a difficult puzzle. Don't just present the solution; walk them through your false starts, your assumptions, and the trade-offs you considered. Articulating your thought process at a senior level means explaining not just what you're doing, but why it's the most efficient or appropriate approach given the constraints.
Embrace Your Inner Quirks (Authentically): Google, and many modern tech companies, genuinely value diverse perspectives. Your unique hobbies and passions can be conversation starters and demonstrate that you're a well-rounded individual who brings more to the table than just code. Be authentic, but also be mindful of how your quirks might translate in a professional setting.
Set Expectations Early and Often: Being upfront about your goals and understanding the company's expectations saves everyone time and potential disappointment. Ask insightful questions about team dynamics, growth opportunities, and even the day-to-day workflow. This demonstrates proactiveness and a genuine interest beyond just the job title.
Failure, The Ultimate Upgrade: After a rejection, don't just wallow. Actively seek feedback if possible (even if it's just reflecting on the questions that tripped you up). Treat each interview as a data point to refine your strategy and identify areas for improvement. Keep a log of questions and your responses to track your progress.

Act III: Bonus Pro-Tips for the Aspiring MAANG Maverick

Scene from Trumbo (2015), directed by Jay Roach

Decode the Interview Matrix: After each interview, jot down the core skills being assessed (DSA focus, system design depth, specific ML concepts). Look for patterns across rounds. Did they emphasize scalability in one and testing in another? Use this intelligence to tailor your subsequent preparation. Think of it as reverse-engineering their evaluation criteria.
Your Recruiter, The Unsung Wingman: Your recruiter isn't just a point of contact; they're your advocate within the company. Build a genuine rapport, ask thoughtful questions (beyond logistics), and keep them informed of your progress and any concerns. A good recruiter can provide invaluable insights and even flag potential areas to focus on.
Become Your Own Digital Detective: While research is crucial, be discerning about your sources. Look for information from current and former employees on platforms like Glassdoor and LinkedIn, but also try to understand the context behind their feedback. Cross-reference information and form your own informed opinions about the company culture and role.
Chase Passion, Not Just the Prestige: Before fixating on a MAANG logo, take a deep dive into your own motivations and long-term career goals. What kind of problems genuinely excite you? What work environment fosters your best performance? Sometimes, a smaller company or a different role might be a better fit for your individual aspirations.
Steer Clear of the "Guaranteed Success" Gurus: While some online resources can be helpful for targeted skill development, be wary of anyone promising a 'secret formula.' The best preparation is personalized and based on a strong understanding of fundamental principles and consistent practice. Focus on building a solid foundation rather than chasing shortcuts.
Mentorship; The Underappreciated Power-Up: Throughout this journey, I realized the immense value of having experienced mentors - individuals who could offer guidance, share their own experiences, and provide honest feedback. If you're navigating this process and feeling a bit lost, seeking out mentorship can be a game-changer. (And if you're looking for personalized guidance, you can learn more about my mentorship offerings at malakarsoham.com/mentorship - I'm passionate about helping others navigate this challenging but rewarding path.)

The Final Word :

As the insightful Beverly Sills once said,

There are no shortcuts to any place worth going.

The path to landing a role at a company like Google requires dedication, resilience, and a strategic approach. While these insights offer a roadmap, your unique journey and your individual strengths will ultimately determine your success. Embrace the challenge, learn from every step, and remember that the 'G in MAANG' isn't just about technical prowess; it's about bringing your authentic self and your passion to the table. Good luck - may your algorithms be elegant and your interviews insightful!

A Leap into Independence: How I bought a car within my first year of graduation

Soham Malakar — Sun, 03 Sep 2023 15:17:42 +0000

The transition from college to the workforce is a challenging and transformative period in one’s life. It’s a time when young adults seek to establish their independence, often marked by significant milestones like securing a job and finding a place to call home. For me, it was also a time when I set my sights on achieving something that felt both ambitious and liberating: buying a car within my first year of graduation.

Setting clear goals

The journey toward purchasing my own car began with a simple yet powerful step — setting clear, achievable goals. I knew that to buy a car within a year, I needed to be realistic about my financial situation and the type of car I could afford. I meticulously calculated my budget, factoring in expenses such as rent, groceries, running costs, and a potential car loan. Having a concrete financial plan was the cornerstone of my success.

Steady job and side hustles

Securing a stable job fresh out of college was a pivotal moment in my life. However, I didn’t solely rely on my primary income to reach my goal. I explored various side hustles, such as freelance, creative work and mentorship to boost my income. The money went into various buckets with a considerable chunk going into my car fund, accelerating my progress.

Frugality

Saving money is often easier said than done, but I was determined to make it happen. I established a dedicated savings account specifically for my car fund. Every month, a portion of my salary was automatically transferred into this account. This strategy ensured that I consistently saved without being tempted to spend the money elsewhere.

Research and negotiation

Choosing the right car was a critical step in this journey. I spent countless hours, days, months researching different car models, their prices, and their running costs — fuel efficiency, fuel variant profitability, maintenance, etc. Once I found a few options within my budget, I was relentless in negotiating the best deal. This involved visiting multiple dealerships, haggling over prices, using various negotiation strategies and being patient until I found the perfect fit.

Practicality over luxury

While it’s tempting to dream of luxury cars, I prioritized practicality and affordability over extravagance. This decision allowed me to stay within budget and avoid excessive debt. My first car though not very flashy, is highly reliable and well-suited to my needs considering the various terrains and distances I have already clocked on this ride.

Managing financing responsibly

To purchase the car, I wanted to pay the whole amount upfront as to not incur the extra interest component of a car loan. My twin brother and I decided to split the expenses, which was one of the best decisions we made in achieving this milestone together.

Celebrating achievements along the way

Every milestone achieved, whether it was reaching a savings goal or securing a better-than-expected deal, was celebrated. These small victories kept me motivated and reminded me of the progress I was making towards my ultimate goal.

Buying a car within my first year of graduation was a significant achievement that symbolized a newfound independence. It wasn’t just about owning a vehicle; it was a testament to the determination, discipline, and financial responsibility. This journey taught me valuable life skills and instilled in me the belief that with clear goals and hard work, even the seemingly ambitious dreams can be turned into reality. As I drove my car off the lot, I knew that this was not just a vehicle; it was a tangible representation of my early successes in the world of adulthood.

Job hunt and lay-offs : Omens

Soham Malakar — Sat, 18 Feb 2023 09:15:01 +0000

It's been a tough start to a new year. Everyone in and around STEM fields have witnessed multiple unfortunate lay-offs happening in every company even the tech giants sitting on mammoth profits. Though the reason for terminating a person's employability isn't quite clear and many are left to wonder that they are incompetent, the truth is far from reality. I'm going to cover quite a few signs to look out for to not fall into the wake of lay-off and even if you're in the wake how to swim out of it.

This article will cover the subtle and blatant signs that you might be a potential Jetsam.

New day same shift

Waking up to a new day, routine fixing bugs of a project you've been working on for the past 10 years might seem like an easy way to mint money but the higher management doesn't. A project with no new updates is very likely to be caught up by a competitor company and realized faster than one might imagine. This is not to be confused with maintaining a software. Maintenance is necessary to make sure the piece of tech works but to make sure people use it, you need to add features, make it look and feel better and to prevent it from becoming obsolete. It's always good to have a talk with your manager to work on something different for a change from monotonicity.

Product/Service is an overhead

Most of us have seen how on-prem products slowly transitioned to cloud. Even among cloud based products we have seen how some products were a hit and some were a miss. There might be a possibility that the product or service you're working on is taking its last breath. It's always better to be informed about the revenue, client acquisition for that product and clients who have shifted from the product. This will help judge if the product is bringing revenue or just burning cash and if it's burning cash, it's better for you to move to a different product/team.

"Work-life" balance has turned to "life" balance

Deadlines are a major part of any project to keep it in line of business. If you find yourself working without hard deadlines or lately your manager has been giving you a slack on your JIRA tasks, certainly that's to prepare you for a smooth and gradual exit. It's counterproductive for a company to keep resources which are availing all benefits but costing the company man hours and profits. It's always recommended complete the tasks on time even if no one is putting hard deadlines.
Disappointed hierarchy

Hearing constructive criticism to improve oneself is good till an extent. However, if you have been hearing from top management constantly to improve your quality of work, then it might be something to be careful about. In the end it's all about the man hours the company is paying its employees for and they would always aim for more work in less time.

Boring tasks

Boring tasks are those kinds of tasks which no one else wants to do or it's not of much use to the company or anyone using it. These are those features which are inconspicuous and of a very low value to a project. The only reason someone might have assigned you this is for you to be prepared for a smooth transition out of the organization. Once in a while boring tasks are fine but if it has been continuing for a long stretch of time, it might be time for you to consider a quick call with your manager.

Conclusion

I have a ratio, which I keep in mind while working in an organization:

The ideal way would be to keep this probability low. So always aim to maximize the results of whatever products you're working on in decent amount of time. There is a way to measure this probability but that is somewhat an involved process, it's time saving to go with intuition on this. Whenever in doubt always make sure to have a one on one meeting with your manager and clear out any confusions which arise.

The complete series is available here. For more content follow here

For mock interviews/resume building/career guidance or any other forms of mentorship click here, to know more.

Towards understanding ML predictions - Introduction

Soham Malakar — Tue, 24 May 2022 17:12:00 +0000

Let's say, we want to quarantine people who have been exposed to a newly discovered virus that has a few symptoms in order to subdue it as soon as possible. However, the symptoms overlap with the common flu virus and we have a Machine Learning (ML) model trained to predict the virus in accordance with the symptom. In order to know what symptoms point to which pathogens and by how much importance, we definitely have to explain the predictions.

Similarly, many a time it might have occurred, why does an ML model predict something, and what might be the triggers for it. For basic models like Logistic Regression or Naive Bayes, explaining the model might be fairly simple but for models like Neural Networks, things get a bit complex.

In this article, we will try to understand how all sorts of models can be explained, without diving into too many technical details.

What is model explainability
Why should we care?
Explainable approaches
Accuracy and intelligibility trade-off
Future Work
References
Footnote

What is model explainability

Model explainability refers to how accurately we can explain the model's predictions or explain patterns for the predictions. For example, let's say we have trained a model to predict if a person has Covid or not depending on 5 symptoms. This can help us determine which symptom is the most potent in classifying the person having Covid.

Why should we care?

For simpler models, we might not have to need to explain the model. However, for more complex models or if we are skeptical about a model's prediction, model explainability can help us.

More often than not, every ML model has a goal to achieve. For certain applications, the goal is to have high recall whereas for certain, high precision. Let's say, you run a model and the P/R numbers are not in the range you were expecting. In order to reach the goal, you need to understand the model.

A few other questions about any model are :

Why did our model make a mistake : Explainability can help us understand the importance of features and how it affects a model
Does our model discriminate : Let's say there is an imbalanced dataset and the model overfits a particular class. For instance, there are many AI-enabled photo post-processing where the AI is trained on Europeans and Americans, so when it does post-processing for any other ethnic group, the photos don't look natural.
How can we understand and trust the model's decision : Every ML engineer has trust issues with his model when it comes to the production environment. Explaining the model helps build that trust.
Are we in a high risk environment : There might be times when there is no room for false positives in predictions, like ruling out if a person has cancer or not? In these scenarios, explaining the model is heavily required.

Explainable approaches

There are 2 ways of explaining any ML model:

Glass box models
Black box explanation

Let us try to understand them in greater details.

Glass-box models

These models are built in such a way that they are inherently explainable. This means no extra steps are required to explain a model after the training the model, all explanations become a part of the model during the training stage itself. In simple words, if we want the explanations we just need to call an inbuilt method that fetches us the explanations.

Since these models are designed to be interpretable, it provides us with exact and accurate explainability for any prediction i.e. no approximations. Using this reasoning how the models make a decision becomes very easy.

A few examples of glass box models are: Explainable boosting machines (EBM), linear models, decision trees, and GAMs

Black-box explanations

Unlike glass box models, black box explanations are made by training a wrapper on top of existing models. In this approach, the wrapper on top of the model tries to explain the model by varying the inputs and outputs of the model.

To understand it further let's take an example. Consider a castle with 2 gates, entry and exit. Both the gates have guards and they are constantly communicating with each other through some medieval telephone. Now, two people - 30 and 10 years old enter the castle in a queue and come out as 33 and 11 years old people respectively. The guards discuss the changes and say that they have grown old by 10%. They further go on to discuss, that the older the people the more they age. Though the example is a bit contrived it clears the concept.

The biggest advantage of this approach is that since it's like a wrapper, it can work on any model. However, its biggest disadvantage is that depending on the varying outputs it gives an approximate explainability, which might be incorrect in a few scenarios.

A few examples of black-box explanations are: SHAP, LIME, global surrogate, PDP

Accuracy and intelligibility trade-off

Let's first define the terms. Accuracy is the measure of how many correct predictions our model is able to provide. Intelligibility is how easy it is to explain the model.

There is this constant trade-off with simpler models having high intelligibility but a lower accuracy and complex models having higher accuracy but lower intelligibility. It is observed in most cases that the simpler models like logistic regression, Decision trees have high intelligibility but low accuracy and it's the exact opposite for complex models like Random Forests and Neural Nets. Now, the question arises of which algorithm to use. We will explore more on this in the next article.

Conclusion

Model explainability is an actively researched area in ML with promising outcomes. It can speed up the lifecycle of ML pipelines by easily explaining the models. In the next article, we will learn more about glass-box and black-box models and how to remove the accuracy vs intelligibility trade-off.

References

Molnar, C. (2022). Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (2nd ed.).
Kübler, R. (2021). The Explainable Boosting Machine.

Special thanks to Yash Govind for his useful remarks!

Footnote

If you liked this article, do let me know in the comments. If you have any questions write me on LinkedIn.

The most Pythonic tools to solve ML problems

Soham Malakar — Thu, 09 Dec 2021 17:14:17 +0000

For any Machine Learning project there is a plethora of python libraries waiting to be exploited and there are also multiple articles in the internet which talks about the usual python libraries such as Numpy, Pandas, Scikit, Seaborn, Matplotlib, etc. However, many a times the basic python functionalities are skipped while reading about those libraries.

In this article we will unleash the full power of some internal functions and libraries of python, which are heavily underrated, for anyone starting a ML project. Without wasting any further time let's directly get down to business.

List Comprehension
PDB - Python Debugger
OS - Operating System
Sets
Time
Venv - Python Virtual Environment
Conclusion
References

List Comprehension

When a list of items needs to manipulated and stored in a different list, or it needs to be manipulated as an intermediate stage before some other operation, list comprehension is a handy tool.

Let's say there is a list and we want to square all the numbers in the list. The usual loop method would be:

list1 = [1,2,3,4,5]
list2 = []
for number in list1:
 list2.append(number**2)
print(list2)

That's a lot of lines and it's time consuming. If we use list comprehension, we can write this in a single line as:

list2 = [number**2 for number in list1]

Cool, now what if we have a 2D list (list of list) and we want to convert it into a list that contains the squares of those numbers?
The normal solution would be:

list1 = [[1,2,3],[4,5,6]]
list2 = []
for data in list1:
 for number in data:
  list2.append(number**2)
print(list2)

A more pythonic solution would be:

list2 = [number**2 for data in list1 for number in data]

You can play around with list comprehension, and you'll never want to use the normal way !

PDB - Python Debugger

This is the inbuilt debugger for python. Let's say there is a scenario where we don't know what is going on with a particular code snippet and it is outputting an unintended result. One way to debug is to put a print statement and print the variables along with some message like:

print('Looks like trouble_1...')

Maybe we need to run the whole code multiple times, before we realize where exactly the issue is. This, will surely delay the project by a substantial amount of time.

The 2nd method is a more pythonic way to debug snippets in a blink of an eye. The PDB, works on the usual principle on how debuggers actually work - by setting breakpoints, and printing call stacks but that's the geeky stuff. The functionality it provides is, after setting breakpoints, one can see all variables at that point in history alongside their values and also create new variables and run codes as they would in a standalone environment.
The way to add the PDB before the suspected error code is given below:

import pdb
# Correct code segment
pdb.set_trace()
# Code here might be a bit sus

OS - Operating System

While running ML codes, there will an urge to store intermediate files and artifacts to certain directories or check even if a directory exists or delete your office's files or run custom shell scripts to hack your neighbors machine, you're going to heavily rely on the os library in python. It almost contains all the methods one is ever going to need to call the Operating System's operations.

Sets

One of the builtin data structures in python is sets. This is very similar to the mathematical set theory. Python sets supports various set operations like intersections, differences, unions, etc.

Sets come in handy when there is some comparison of data or finding unique entries in a file or extracting common entries in data or doing some extraction operations on data. Let's take an example, suppose there is are 2 sets:

fruits = {tomato, apple, banana, orange}
veggies = {tomato, cabbage, potato, onion}

Now, to find out which food item is a fruit as well as a veggie, we can easily do a set intersection like so fruits.intersection(veggies)

If we wanted to do this with the usual way, the most naive way would we would have to run 2 loops and compare the elements and take another list where we have to keep appending the common food items.

Time

An ML Engineer's most important resource is time and there might be times where a script is taking way long to run. There can be performance issues with the code for various reasons. Before figuring out which part of the code is taking the longest it'll tougher to pinpoint the issue. For pinpointing the locations for longest running snippets, the time library plays an important role.

Venv - Python Virtual Environment

This one is the most important library which every ML engineer uses. It creates an independent environment where one can run their scripts, this eliminates multiple dependency issues in a project.

Let's try to understand this with a scenario. Suppose, there are 2 projects and both of them require different versions of a python library in order to run with a constraint that at a point in time there can only be 1 version of the library installed. This looks impossible to run both scripts on a single machine, which is true. There are many ways to solve this problem, the simplest way is by creating 2 different environments and installing the required dependencies in the respective environments and then running the scripts in their respective environments.

As a ML Engineer, there will be multiple projects to work on simultaneously and it's always recommended to use a different environment for different projects so as to not run into any dependency issues.

Conclusion

These are a few tools I realized were basic, powerful yet underrated for a person starting any ML projects. These few tools will not only boost your productivity but also make you realize why python is the de facto language for ML projects !

References

https://docs.python.org/3/library/pdb.html
https://docs.python.org/3/library/os.html
https://docs.python.org/3/library/time.html
https://docs.python.org/3/tutorial/venv.html

Parkè

Soham Malakar — Tue, 09 Nov 2021 19:52:16 +0000

Whether you are moving across town or towards your dreams, having a safe reliable spot for your transport brings a huge peace of mind. Let us have your snags. Reserve your space, drive sans worries.

Parkè is an idea I along with a few other friends had conjured in our junior year of college, because of a common problem we were facing; parking our vehicles.

I am going to divide the whole topic into the following sections:

Present scenario

Not only is driving loved around the world but also hated for the lack of proper parking spaces. All of us at some point have felt the need for a regulated and centralized parking management system while driving.

While finding a parking spot takes away huge chunk of our time, on reaching we either find it full or comes the long waiting until cars are rearranged to create bumper to bumper spaces with unreasonable rates. Making parking a tiring, tedious work with your vehicle susceptible to regular paint damages if not dents. In addition to that, many spots have a disclaimer stating that parking is at owner's risk.

The solution to all these problems is Parkè.

Features of Parkè

Buying you time

We bring a hassle free experience at the ease of your fingertips.

Search for spots around you, select the spot which best suits your needs. We guide you through the fastest route, on reaching, the space is all yours. It is as simple as that!

Safety and security

Your vehicle is safe with our advanced safety and security measures in every verified parking spot. Fixed number of parking spots make sure every car has ample padding spaces

Passive income

One can apply for hosting their space on Parkè. After we verify, it can be up and provide them earning from this space.

Regularizing parking practices

Unregulated parking practices is an unavoidable part of our daily lives.

No proper receipts, charging more than the rate and so on. At present date, there are no measures to monitor and stop the suspicious middle men and their questionable practices.

Parkè ensures a convenient and cashless way to counter these issues by conducting transactions directly between customers and Parkè without involving any 3rd party.

Prototype of Parkè

There are 2 flows to this:

Finding a parking spot (Customer)
Listing a parking spot for others to park (Partner)

Both of these are pretty straight forward as our application guides you on how to set up the flows. Talking of application, you can choose to download the native application or use the web application to complete the flows. For more info you can visit the website.

Next is the overview of our native Customer application

And here is the overview of the native Partner application.

Nothing's better than a video showing the prototype in action right? So here's the clip of Parkè in action.

Challenges faced while developing Parkè

Scalability

In the back-end, our biggest challenge was to choose the appropriate database for our use case, because database is one of the most important things in an internet based application. We chose MongoDB, as it is highly scalable for a huge user base.

Another issue we faced was regarding the fetching and showing of the nearest parking spots to our customers. Our 1st plan was to fetch all the data and find the distance between those coordinates and the customer's current location, but that was taking too much time. On further research, we implemented the same by tweaking a bit of MongoDB's Geo-spatial features.

For the database, we have different tables for storing the customer and partner data. If it would have been in the same table then fetching the data for a particular entity would be more resource intensive as it would have to search longer. It will also be easier to analyze the data while doing some business scaling.

Space

With respect to the storage of a variety of vehicles like bikes, hatchbacks, etc, the problem we faced was with the absolute spaces the vehicles take. Since the bikes, hatchbacks, etc do not have a definite size comparative ratio the only option was to physically go and check the parking spaces, which would be a drain of revenue from a business perspective. Therefore we came up with a solution of keeping the space constant for all vehicles as done in public parking spots.

Safety

As Safety is our highest concern, in order to validate the RC of the vehicle of the customer we are using our own custom API which checks the details from a trusted source and verifies the car.

Ease

Slack time of 10 mins is given between booking and reaching to prevent chaos at parking spot.

It could have been a common issue of traffic congestion if right after the spot became empty, it would show in the vicinity. To check this we have kept a 30 mins cool down for a smooth traffic flow.

Optimization

From a technical perspective, with the multitude of people using our application, it would be unwise to fetch consumer details every time the application ran. Therefore we cached local data to prevent repeated fetches.

Expansion plans for Parkè

To provide premium memberships enabling access to exclusive parking lounges with services and features such as EV charging ports, car washes, servicing, repair and maintenance options and accessory outlets.

Setting up EV charging infrastructures throughout could be a major push to see more EV vehicles on the road and pave a road to a safer and cleaner future.

Blockchain integration for payment as this might be the future of currency exchange.

Why don't we hear about Parkè

Very intuitive question indeed, I mean if the idea is so good, we should be as famous as any on-demand transport company such as Uber.

The most important reasons why we couldn't pilot it as we expected :

Lockdown due to Covid-19

We were working on this idea remotely, therefore things took a bit longer than it should have. People were traveling less and less traveling meant less active users, empty parking spaces, less revenue for partners. Overall, there was a less user count.

Minimal funding

Since our active member count wasn't very great due to lockdown, it was tough getting funded as we didn't have good metrics to provide to investors.

Lack of mentorship

We were a bunch of college students, with not much knowledge about starting a business or running an organization.

Co-founders of Parkè

Complete Interview Experience: Informatica

Soham Malakar — Mon, 13 Sep 2021 17:03:12 +0000

Around the month of April, Informatica had visited our University for hiring Interns.

There was a total of 6 rounds in the whole selection process:

Online Aptitude Round
Online Coding Round
1st Technical Round
2nd Technical Round
3rd Technical Round
HR Interview

Online Aptitude Round

A total of 300 students were shortlisted for this round.

In this round we had to solve Multiple choice questions pertaining to computer fundamentals such as: C programming language, Database Management Systems, Operating Systems, Data Structure and Algorithms. There were 30 questions to be solved in 45 minutes.

Online Coding Round

Out of the initial 300 students, only 29 were shortlisted for this round.

We were given 4 questions to solve which most likely resembled:

1st Technical Round

This was the first face to face real-time interview held the next day from 9am in the morning. I was asked about the details I had provided in my resume.

Following which I had to share my coding screen with the interviewer. They asked to code on puzzles related to Stacks, Linked List, Recursion and Binary Trees. There was no constrain on the programming language and it was the discretion of the interviewee to select the programming language. Around five to six coding questions were asked.

Since I was coding in Java, the interviewer asked me questions on Java fundamentals. After which general computer fundamentals questions were asked on topics learnt in B. Tech course.

This interview lasted for 45 minutes.

2nd Technical Round

This was second face to face interview held on the same day at 3pm.

After summarizing my resume, I was required to share my coding screen again for testing my technical skills. I was asked some advanced questions on advanced topics of Linked List and Trees. There were also a few puzzles which took some time to solve. Around five questions were again asked in this round.

Following which, we moved on to the computer fundamental round, where I was asked questions on Operating Systems, Database Management Systems and Data Structures and Algorithms.

This round lasted for around an hour.

3rd Technical Round

Around 5pm, the third face to face interview was scheduled.

Similar to the previous two rounds, first I was supposed to summarize my resume in 2 minutes. This time I had to share my whole screen to the interviewer. I was asked around 3 coding questions related to Sorting, Searching and Heap.

Following which I was asked fundamental questions on C and Java. A few of the questions were also comparative in nature.

HR Interview

Finally, around 7pm, I had my HR interview.

As expected, this was completely different from the Technical Rounds. Over here I was asked subjective and existential questions as to determine what kind of person I am. My resume was meticulously scanned and I was asked questions on the details provided on it.

The round ended around 8pm.

Results

The next day around 12 noon I received a mail inviting me to join an online video conference to congratulate the interns who were selected as Interns.

Finally, I got selected as an Intern for Informatica !!

Securing an Internship in IIT

Soham Malakar — Sun, 12 Sep 2021 14:01:52 +0000

Securing an internship in one of the most prestigious universities of the county is a dream for many students and it’s no different when it comes to the IITs. In the college years of an engineer, at one point or the other he yearns for an internship in any one of the IITs. Today, I’m going to enumerate a few steps by which you can get an internship in one of the them. I secured an internship opportunity at IIT Kharagpur, I’m sure that you can do it too.

The procedure for internship is more or less the same for any IIT so the points are going to be quite general.

1. Calendar

Different IIT’s have different periods during which they take candidatures for the summer internships. You need to be aware of these dates because once the form expires the opportunity is missed.

2. Skill Hunting

Since these are mostly research internships, you need to acquire some skills and need to have an idea about the domain you are passionate about, before applying for the internship. You can expect the internship to be on some government projects or industrial projects. The professors are very kind, they will provide you all the necessary help you require, once you are selected and the internship starts. So, there won’t be any training as such because this is not industrial internship.

3. List

There are so many professors in so many IITs with eclectic research domains, short-listing them seems like a mammoth task. Well, it is a mammoth task, so what you can do is, make a list of professors consisting of:

Name of the professor
Research Domain which are more or less similar to your skillset, don’t be very choosy
The contact details of the professor

You can find all these details in the IIT websites itself and make sure to complete this list before the dates when the forms open.

4. Communicate

This is an optional step, but this increases your chances to get an internship. When the time is perfect and the forms are open, send a well composed mail to the professor stating about:

Yourself
Your skillset and certifications
Your Achievements
Why you want to do the internship

Many people think that writing their GPA or CGPA increases their chances of getting an internship; personally, I feel that marks is not a very good metric to evaluate someone’s knowledge. Instead, you can write about the projects you have worked on related to those domains and give a link to your GitHub and LinkedIn profiles.

Again, you are free to structure your mail according to your liking.

5. Wait

After following all these steps, you need to do the toughest of all steps, wait for a response. If you have the required skills and a few projects to showcase, you will surely get a reply sooner or later. It’s also quite natural for the professors to not reply as they are getting a lot of mails and they are humans, so don’t get disheartened.

6. Internship at IIT

Congratulations on getting an internship in IIT. While you are doing your internship, don’t forget to enjoy and live the experience, I can’t emphasize enough on how important that is.

7. What if you don’t get an internship in IIT?

Firstly, you tried, where most people were busy enjoying their time so kudos to you on that. There is nothing to get morose about, you just need to develop your skills a bit more and add a few more projects to your portfolio. It’s all a learning process and failure is the best teacher.

Conclusion

If you found this article useful, drop a react and share it with your near and dear ones. Good luck for your journey ahead!

Detecting Welding Defects in Steel Plates using Computer Vision Algorithms

Soham Malakar — Sat, 11 Sep 2021 14:05:17 +0000

Introduction
Prerequisites
Image Segmentation
Image Moments
Understanding the Data
Method and Algorithm Used
Results
Conclusion
Co-Authors
References

1. Introduction

Welding defects can be defined as weld surface irregularities, discontinuities, imperfections, or inconsistencies that occur in welded parts. Defects in weld joints could result in the rejection of parts and assemblies, costly repairs, significant reduction of performance under working conditions and, in extreme cases, catastrophic failures with loss of property and life.

Furthermore, there are always certain flaws in the welding due to the inherent weakness in welding technology and the characteristics of metals.

It is important to evaluate the weld quality, as welded joints are often the locations of crack initiation due to inherent metallurgical geometrical defects, as well as heterogeneity in mechanical properties and the presence of residual stresses.

In practice, it is practically impossible to obtain a perfect weld and, in most circumstances, it is not necessary to provide the adequate service functions required. However early detection and segregation is always preferred to mishaps.

Using our algorithms, we can easily detect faults in welding by the images and also measure the severity of each fault precisely. This will further help in faster image recognition and avoid adverse situations from arising.

It was found, using Convolutional Neural Networks algorithm and U-Net architecture making the process much more efficient. Resulting in an accuracy of 98.3% by the end of the work.

2. Prerequisites

Basic understanding of Machine Learning
Basic idea of Convolutional Neural Networks
Understanding of Convolution, Max Pooling and Up Sampling operations
Idea of U-Net architecture
Basic understanding of skip connections as in residual blocks (Optional)
Working knowledge of ConvNets with Python, TensorFlow and Keras library (Optional)

3. Image Segmentation

Segmentation partitions an image into distinct regions containing pixels with similar attributes. To be meaningful and useful for image analysis and interpretation, the regions should strongly relate to depicted objects or features of interest.

The success of image analysis depends on reliability of segmentation, but an accurate partitioning of an image is generally a very challenging problem.

A chest x-ray with the heart (red), lungs (green), and clavicles (blue) are segmented

4. Image Moments

An image moment is a certain particular weighted average of the image pixels’ intensities. Image moments are useful to describe objects after segmentation.

Simple properties of the image which are found via image moments include:

Area (or total intensity)
Centroid
Information about its orientation.

5. Understanding the Data

The dataset contains two directories. The raw images are stored in images directory and the segmented images are stored in labels directory.

Let’s visualize the data:

Raw Image from images

These raw images above are RGB images which have to be used for training the model as well as testing the model. These pictures vary in dimensions. Intuitively, the darker portions are welding defects. The model needs to perform Image Segmentation on these images.

Binary Image from labels

These images from labels directory are binary images or ground truth labels. This is what our model must predict for the given raw image. In binary images the pixels have either a high value or a low value. The white region or the high values denote the defective regions and the black region or the low values denote no defect.

6. Method and Algorithm Used

The architecture we are going to use for this problem is U-Net. We are going to detect faults and measure the severity of these welding images in three broad steps:

Image Segmentation
Showing Severity using Colour
Measuring Severity using Image Moments

Training the Model

The following is the U-Net architecture we used for our model:

The U-Net Architecture used

Points to Note:

Each blue box corresponds to a multi-channel feature map
The number of channels is denoted on the top of the box.
The (x,y) dimensions are provided at the lower left edge of the box.
The arrows denote different operations.
The name of the layer is provided below the layer.
C1, C2, …. C7 are the output layers after Convolutional operation
P1, P2, P3 are the output layers of Max Pooling operation
U1, U2, U3 are the output layers of up-sampling operation
A1, A2, A3 are the skip connections.
The left-hand side is the contraction path, where regular convolutions and max pooling operations are applied
The size of the image gradually reduces while the depth gradually increases.
The right-hand side is the expansion path, where (Up Sampling) transposed convolutions along with regular convolutions operations is applied
In the expansion path, the size of the image gradually increases and the depth gradually decreases
To get better precise locations, at every step of the expansion we use skip connections by concatenating the output of the transposed convolution layers with the feature maps from the Encoder at the same level: A1 = U1 + C3 A2 = U2 + C2 A3 = U3 + C1
After every concatenation we again apply regular convolutions so that the model can learn to assemble a more precise output.

Model is compiled with Adam optimizer and we use binary cross entropy loss function since there are only two classes (defects and no defects).

We use a batch size of 10 with 100 epochs(The number of times the model is run on all inputs).

Please note that there could be a lot of scope to tune these hyper-parameters and further improve the model performance.

Testing the model

Since the model takes input of dimension 512x512x3, we have resized the input to that dimension. Next, we have normalized the image by dividing it by 255 for faster computation.

The image has been fed to the model for predicting the binary output. In order to amplify the intensities of the pixels the binary output has been multiplied by 1000.

The image is then converted to 16-bit integer for easy image manipulations. After which an algorithm detects the defects and visually marks the severity of the defects through color grading along with assigning weights to the pixel with defects depending on their severity. The Image Moment is then calculated on this image considering the weighted pixels.

The image is finally converted back to 8-bit integer and the output image is displayed with color grading and its severity value.

7. Results

The visual metric we are using for severity detection is colours

In the image, the colours:

Green signifies areas with severe defects.
Blue corresponds to areas with more severe defects.
Red areas show the most severe defects.

The 0th moment is displayed as a percentage alongside the output image as an empirical measure of severity.

The following are three random samples which shows the raw input, the ground truth and the output generated by our model.

Sample 1:

Raw Image

Binary Image (Ground Truth)

Predicted Output with Severity

Sample 2:

Raw Image

Binary Image (Ground Truth)

Predicted Output with Severity

Sample 2:

Raw Image

Binary Image (Ground Truth)

Predicted Output with Severity

8. Conclusion

Thank you for interest in the article. Please leave your valuable feedback and suggestions 😃

Full code on my GitHub repository here.

9. Co-Authors

Soham Malakar — Connect with me on LinkedIn, here

Swastik Malakar — Connect with me on LinkedIn, here

10. References

https://www.cs.auckland.ac.nz/courses/compsci773s1c/lectures/ImageProcessing-html/topic3.htm#adaptive

https://medium.com/r/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FImage_moment

https://medium.com/r/?url=https%3A%2F%2Ftowardsdatascience.com%2Funderstanding-semantic-segmentation-with-unet-6be4f42d4b47

https://www.sciencedirect.com/topics/materials-science/welding-defect

https://domingomery.ing.puc.cl/material/gdxray/

Why Creative people are like Coriander

Soham Malakar — Fri, 10 Sep 2021 12:37:49 +0000

Have you ever felt like an apple in an orange orchard? Have you ever felt like a fountain pen among quills? Have you ever felt like the sober person among sloshed friends?

Well, then I must say we share relatable snippets from our life’s canvas, most of which is inked with estrangement.

Allow me, to sketch my life’s timeline while you read this and try to relate. It all began from the time we stepped through the threshold of our school, we had friends, many of them. But as time progressed we disintegrated into small groups of our own, groups with same wavelength of thought with traits that were similar. Nobody, our parents, teachers, seniors, nobody, told us to polarize, but I guess it’s a natural human tendency to do so. As long as one was in a group he had a fraternity to back him up, and the person who lacked a group had the support of his own isolation and sagacity. As we got older, so did our groups, and the brotherhood strengthened.

Now, we were in middle school and talking about other people and groups had become a norm. But as voluntary solitude is very different from communal solitude, so was the demeanour of the people practicing it. The voluntarily isolated people were considered outlaws and nobody really talked to them because they always kept a distance from the majority but the latter kind of people were like coriander, not particularly required in a dish but goes well with most of the dishes. They had the potential to talk with everyone in their batch but like a desolate guy’s fate would have it, no one in their batch would take the first step to talk to them, instead they had to initiate the conversation, always.

I bet most of you would agree when I say that they were the real pioneers who actually did new things, creative in their work, who thought way more and way deeper than the hoi polloi. Now, one might say that everyone would like to be friends with such a person, well, yes, but when someone outperforms others of the same age group, same batch; green eyes shine brilliantly and what follows is perpetual silence from the fellow competitors, not the curious.

We were in high school and time flew like an arrow, steady but swift. We found ourselves standing at tapering end of our school life, Mr. Corianders had realized by now that they had to live their life in the present way throughout, since the price one pays for being different from the crowd is ostracism.

College came as a reassurance to the conviction already formed. This was the time when they tried to gel in with the people but in that way, they began forgetting their true identity. After spending some time in that futile attempt to gel in, they realized that it was just not their cup of tea. Now, one might think life must have been very tough living in solitude but, knowledge set them free. Existentialism, it was, and a few concepts of higher physics that mostly dealt with philosophy. Following what the wise people had set as a precedent, they started being more of themselves and not the person complying with the specifications and rules others had set for them. Thus, began another stage of metamorphosis, self-realization.

Thoughts of “to be or not to be”, were flushed in the wake of self-realization

The urge to appease others in order to be a part of the crowd had dampened. Everything headed the way it was before, the whole idea of being apart from the crowd, being an apple in an orange orchard, being a fountain pen among quills, being the sober person among sloshed friends.

If you, dear reader, find this an exceedingly relatable walkthrough, and are in search of some motivation, just remember —

What is alienated today due to inability of the society to comprehend, might be the driving force of tomorrow. Be the bulb among candles!

Yours truly,
Mr. Coriander.

How Quarantine changed my life !

Soham Malakar — Tue, 07 Sep 2021 15:52:51 +0000

Hello, ladies and gentlemen!

I am presuming that your curiosity led you here to solve the mystery behind the title.

Or as Frank Herbert would put it,

The mystery of life isn’t a problem to solve, but a reality to experience.

Tighten your seat-belts for here we go on a sequestered adventure.

It was spring, flowers, honeybees, butterflies could be seen welcoming the refreshing season. However, it wasn’t a very good time for us human beings. Men as we are, clumsy in our ways, we had somehow unleashed a dreadful virus which called for measures which was going to take away the quintessential reason we were called humans, our social life.

It was no different for me. My college semester exams were over so I had taken a short leave to visit my home. Little did I know that, it was the last time I was seeing my hostel, and friends in college. I packed my bags and left with joy and happiness. After a few days, college sent me a mail stating:

All college functions will be shut down due to Corona Virus, except the essential services, till further notice

Moreover, our statesmen had also put up quarantine and home isolation in order to prevent the spread of this lethal virus. Which meant no socializing, hitting the church of iron or visiting the nearest restaurant.

Initially I was happy, because, more time in home meant, good food, unrestrained sleep at odd hours, video games, Netflix, and so much fun, right?

These were a pretty good deal…till I got bored of all the joie de vivre. There was something telling me that I should be up and doing something that is out of my comfort zone. A spend a huge amount of time thinking about how possibly I could spend my time better.

Pretty tough question, right? That’s why I asked two of my friends about their opinions regarding doing anything unconventional this summer a.k.a. quarantine. Needless to say, I had put them on an eternity of contemplation…

However, I thought it was the best time to do home workouts because let’s face it, most, probably all of us have tried home workout challenges at some point of time but it didn’t really work out. So, I took thirty days of a hundred push-ups challenge with my friend and to my surprised I actually completed it. Truly it was more of challenge to my mental strength than my physical strength and trust me, you are your best project. Thus, I carried on this fitness regime because now it had become a lifestyle.

Humans, being social animals, it is quite tough to just stay indoors and not interact with people. It was getting pretty uncomfortable for me too, just sitting indoors and watching the birds interact. While accepting the fact that we made a mistake for which we had to be socially isolated but as humans we can’t disregard the fact that we are smart too, we invented technology. So, I decided to talk with random people, complete strangers on social platforms to connect with them or as I would put it, “virtually socialize”.

Frankly speaking, I didn’t expect anyone to respond but around six people responded out of thirty-ish conversations I had started. It was all good and honestly I felt my inter-personal skills had upgraded until they stopped talking after a point of time, which is a personal choice and I respect that. Normally people don’t just stop talking so I began pondering about what the possible reasons might be because explicitly asking them was an exercise in futility. There was this common thing I noticed about that “threshold” and it was “personal gain”. Since the conversations were completely friendly and had absolutely nothing to do with money or fame, the messages were seen with no replies to expect. It appears that people have become more materialistic and have devalued human touch to a greater extent. It wasn’t really how I had planned it to be but it boosted my confidence to talk with strangers.

Life was still good until we had a major tropical cyclone in West Bengal and Odisha — Cyclone Amphan. Never did I feel so daunted by a difference in air pressure until now. While we were still recovering from this crisis, another news came about locust attacks in Northern India. Biblical verses never seemed so real, the locusts were literally in swarms which covered few ten kilometers and not only destroyed crops but also trees. Food crisis was now an inevitability.

What troubled me most was the plight of the people who were not as privileged as me. At least I had a roof over my head, food stored in my refrigerator, and fresh water to drink — the basic necessities of any human being, but what about those people who had lost their accessibility to these basic needs, how would they survive?

Leaving aside the physical aspects of life, people had also suffered psychologically with so many people losing their near and dear ones because of the deadly virus, people losing their priced possessions and there is nothing worse than a broken spirit. These were tough times when people required mental support. But its a human fallacy to stay out of anything that’s not unicorns and rainbows just out of fear of the other person getting dependent on them. Thus, its one of the utmost virtues to have gratitude and respect for those who stand by our side at our time of need.

A lot of international news were also making headlines recently. One of which was the brutal homicide of George Floyd by the state’s keepers of peace. This had garnered international support with protests happening all over the world along with larceny and arson. People took this as an opportunity to expose everyone they thought were guilty of using “White Privilege” or promoting racism, ranging from actors, statesmen(old and new), public servants, sportsmen, etc. I would refrain from getting too political on this topic but I would like to quote a line from Sapiens: A Brief History of Humankind —

Unfortunately, the Sapiens regime on Earth has so far produced little that we can be proud of. We have mastered our surroundings, increased food production, built cities, established empires and created far-flung trade networks. But did we decrease the amount of suffering in the world?

Just as I was writing this article, I received an unfortunate news about Mr. Sushant Singh Rajput, one of the best actors of Bollywood, had committing suicide for an underlying condition of depression. The reason I bring this up is because I feel people very conveniently use “depression” interchangeably with “sadness”. Depression is more deeply rooted than sadness, yet it is curable.

The real issue is, society underrates male depression and tells them to “man up” for whatever the cause may be.

I’m awestruck to see people who ignore and seen-zone messages, all year round, suddenly start caring through their well-crafted social media posts whenever it concerns a celebrity. But again, that’s the deception and the hypocrisy of the society we live in.

To sum up, I have made points of my realizations of the above worldly events:

The best project you will ever work on is, yourself
Don’t take anything for granted — food, people, friends
People are materialistic
Don’t waste what you have, it might be a privilege and luxury for some
Have gratitude and respect for those who stood by your side at your time of need
Don’t consume everything an organization sells, think for yourself
Save for the future
If we live at the cost of others, we fail as a society There’s nothing worse than suppressed emotions
Life goes on

Why October is so important

Soham Malakar — Sat, 07 Nov 2020 08:02:01 +0000

This year, October was a pretty busy time for developers, tech enthusiasts, students, engineers and of course Charlie and his chocolate factory. Needless to say, as developers we know how valuable stickers and T-Shirts are in the tech community. If a developer was to choose between a bottle of the finest wine and a bunch of stickers, he would choose the latter.

With all that said we will focus more on the technological part of October without completely ignoring Charlie.

HactoberFest

This is probably the most famous tech fest that is well known among developers and students alike. It is a month-long celebration of open source software aimed at introducing beginners to Open Source Community, brought to us by Digital Ocean, Intel and DEV Community.

This is a great event to implement your learning on real life projects and helping the community build something useful.

Google DevFest

As they would put it ‘The Code Concert’. This is an event where global developers come together for the weekend and share their technical learning and expertise on Google Technologies. An initiative of Google, Google Developer Groups, Women Techmakers and Google Developer Experts, this event empowers developers to teach, learn, and connect with other developers.

This is a great event to learn new technologies or brush up on basics and learn new ways to implement tech. Since in the tech world everything is interdisciplinary so learning one technology empowers one to understand all other similar technologies.

// Detect dark theme var iframe = document.getElementById('tweet-1322672892719452160-228'); if (document.body.className.includes('dark-theme')) { iframe.src = "https://platform.twitter.com/embed/Tweet.html?id=1322672892719452160&theme=dark" }

#30DaysOfGoogleCloud

This is a more like a free coding bootcamp to give people an idea about various Google Cloud services and how/where to implement them with hands-on training. Generally, all the companies offer similar Cloud services so learning any one gives you control over the others too.

// Detect dark theme var iframe = document.getElementById('tweet-1344575389163692032-133'); if (document.body.className.includes('dark-theme')) { iframe.src = "https://platform.twitter.com/embed/Tweet.html?id=1344575389163692032&theme=dark" }

Halloween 🎃

As promised, here’s Charlie. Charlie knows that all the developers are exerting themselves to learn and implement their technical knowledge in the various events discussed above. Charlie also knows that chocolate and celebrations are a way to ease stress and reduce burnouts. That’s why Charlie makes a variety of chocolates and candies in his factory for the event that marks the end of October.

Charlie is smart that’s why he wants everyone to celebrate festivals and take some time out to rest and reward yourself.

Conclusion

These swags are just incentives. While participating in these events don’t forget what the idea behind that event was else that just destroys the purpose. So, while you plan your next eventful month, I will make myself some ☕

DEV Community: Soham Malakar

Is the LeetCode Grind Overrated? My Unexpected Google Interview Success Story

Act I: The Interview Gauntlet

Act II: My Semi-Serious, Mostly Sarcastic Survival Guide

Act III: Bonus Pro-Tips for the Aspiring MAANG Maverick

The Final Word :

A Leap into Independence: How I bought a car within my first year of graduation

Setting clear goals

Steady job and side hustles

Frugality

Research and negotiation

Practicality over luxury

Managing financing responsibly

Celebrating achievements along the way

Job hunt and lay-offs : Omens

New day same shift

Product/Service is an overhead

"Work-life" balance has turned to "life" balance

Boring tasks

Conclusion

Towards understanding ML predictions - Introduction

Contents

What is model explainability

Why should we care?

Explainable approaches

Glass-box models

Black-box explanations

Accuracy and intelligibility trade-off

Conclusion

References

Footnote

The most Pythonic tools to solve ML problems

Contents

List Comprehension

PDB - Python Debugger

OS - Operating System

Sets

Time

Venv - Python Virtual Environment

Conclusion

References

Parkè

Contents:

Present scenario

Features of Parkè

Buying you time

Safety and security

Passive income

Regularizing parking practices

Prototype of Parkè

Challenges faced while developing Parkè

Scalability

Space

Safety

Ease

Optimization

Expansion plans for Parkè

Why don't we hear about Parkè

Lockdown due to Covid-19

Minimal funding

Lack of mentorship

Co-founders of Parkè

Complete Interview Experience: Informatica

Online Aptitude Round

Online Coding Round

1st Technical Round

2nd Technical Round

3rd Technical Round

HR Interview

Results

Securing an Internship in IIT

1. Calendar

2. Skill Hunting

3. List

4. Communicate

5. Wait

6. Internship at IIT

7. What if you don’t get an internship in IIT?

Conclusion

Detecting Welding Defects in Steel Plates using Computer Vision Algorithms