Technical Interview Questions

Samarth S. - "Before we jump into discussion , few Qs which come to my mind in order to narrow the scope are: What kind of security system we are thinking . What do we want to achieve? Is there any biz goal in mind. What security related pain points are we trying to solve. What is the kind of hotel - is it a vacation rental , standalone hotel, highway motel, big chain resorts? What are the current systems being used. I think most of hotels have closed circuit cameras , fire alarms installed in"See full answer

Anand K. - "I started off with defining the users of the system. From them i dervied the functional requirements and categories into MoSCoW. Then for the DB part, Userprofiles, App data to be stored in a RDBMS. The binary apps to be stored in a scalable, distributed DB such as S3. The app data from RDBS to have a link to S3 URLS. Then i touched upon Partition tolerance while scaling, Replication, Caching at multiple levels, Load balancers at multiple levels. This being my first System design interview i"See full answer

Hmm T. - "Below are few key things enterprises look for in a data center : Location Proximity to target user base Potential risk exposure due to natural calamities (hurricane, storms, tornedos etc) Real estate cost (Service cost will also be proportional to how expensive it is to acquire and run data center) Security Physical security Logical security (Segregation) Industry standard compliance PCI DSS ISO etc Ability to support HA and DR requir"See full answer

Exponent - "One technique we like to use when explaining technical concepts to non-technical people is to use analogies to help with memory retention. For this type of question, we'll use the Restaurant example: > Imagine you're at a restaurant with a variety of waiters - each waiter will only take your order for one specific item (i.e. one waiter for pizza, one waiter for pasta, etc.). An API is like a waiter in this example as it will only perform a specific order. An SDK on the other hand is like having"See full answer

Musonda C. - "TCP and UDP are ways in which data is transferred over the internet. TCP needs to have a connection between the two hosts in order for data to be transferred, while UDP begin the transfer without waiting for a connection between the two hosts. This makes UDP faster, but not as reliable as TCP because there is a chance the data transfer wont happen in UDP due to the uncertainty of the connection. TCP is used where reliable data transfer is crucial e.g file transfers, email, web browsing and UDP"See full answer

Promise S. - "Submitted before deadline and described the creativity behind the new website"See full answer

Karthik M. - "Assumptions & Clarifications Users have access to all Google Workspace applications like Google Drive, Gmail and the same photo editing tool - let us call this tool - Google Photo Editor (GPhoto). Users can share the images like Google Docs - giving people read-only, read-write / edit access Goals Google's vision is to organize the world's information and make it accessible. GPhoto's mission / goal is to organize the world’s photo and make it accessible to others to view, edit or"See full answer

Sadhana U. - "Know your audience Understand who the end user is and their role within the organization. Adapt technical lingo to lingo that can be understood by the end user. Diagrams of functions with less dense text is helpful for non-technical users. If it presents like a Manual, that's a turnoff. If it presents like a Guide, non-technical people can follow it. Simplify the problem By simplifying the problem you can explain it to your client in her/his own words. Try to use examples from the client'"See full answer

Adib M. - "Manage Programs that are in flight Track the progress vs Plan and see if things are on track, falling behind Monitor Risks and update current and any new risks identified and mitigation approach Communicate updates to stakeholders (weekly/biweekly as agreed) Communicate status to peer TPM and partnering teams Daily sync with Engg (like a timeboxed daily Standup to understand Daily Plan and any impediments raised that need resolving) Facilitate Reviews with Product Manager as features"See full answer

Parth S. - "Assuming that trades will have information like trade_type buy or sell trade_price with these tuples, one can iterate over each trade while maintaining a stack which maintains all the open buy trades. If we encounter a sell trade then we pop one element make it a buy/sell pair and calculate the profit/loss for that pair. Moreover, keep adding pair-wise profit/loss to calculate overall profit as we continue iterating over trades. At the end print pairs and their profit/loss along with"See full answer

Praful B. - "Define the goal of the experiment. What do you want to test? Are you trying to improve conversion rates, increase engagement, or something else? Identify the variables to test. What are the different options that you want to test? For example, if you are testing different call-to-action buttons, your variables would be the different button text and colors. Set up the experiment. This involves creating two versions of the app or website, one for each variation of the variable that you are testi"See full answer

Seamus L. - "The first part is to design with either a linked list of IDs or an array of boolean values. The second part is to optimize both time and memory."See full answer

Tiffanie K. - "licensing violations for mis-use, integrity of the data (how often is data refreshed), unstructured/garbage data, potential to download virus, mis-interpreted data, unreliability of system (i.e. 3rd party unplanned downtime),"See full answer

Tj C. - "SELECT (zip, rides, deliveries) FROM dataset GROUP BY zip;"See full answer

Marwan O. - "My company was acquired and I was responsible for planning and managing a transition & integration of custom real-time time-series data and event capture software. I worked with lab scientists (who were running the instruments from which the data would be captured) to understand their unique needs compared to the use cases the software was originally written and deployed for. I wrote proposals for integrating specific components of the software and abandoning others, receiving review and a"See full answer

Surbhi G. - "In order to train a machine learning model for license plate detection, which is essentially image recognition, I would first need a large labeled dataset of images featuring license plates. From hereon, I would use a model architecture suitable for image-related tasks, like CNNs (Convolutional Neural Networks). When it comes to deploying the model, I would first optimize it, aiming for the best balance between speed and accuracy. Depending on the specifications, the model could be deployed eith"See full answer

Parth P. - "Establish more servers on regional bases or locally and move network traffic to these servers instead of keeping on one server."See full answer

Surbhi G. - "My preference would likely be server-side deployment given the specification of millions of users. However, it is worth considering a hybrid model too; where basic, non-sensitive, and less compute intensive tasks are processed on the client-side, while the heavy-duty processing is done on the server side. The main advantage of client-side deployment is it can work in real-time and offline. It also significantly reduces the load on your servers, making it more efficient in handling other tasks. N"See full answer