System Design Q&A and Videos

For my current project, I have 2 broad options for the infra/platform to support the feature:

1. Using a proven technology that has existing use cases within Meta but a somewhat limited API
2. Using a new technology that's not completely done yet and has uncertainty, but should have better APIs and performance

There's also the option of trying to do both, benchmarking them against one another, and then choosing the better one. However, I don't think we'll have the time to do that.

All this being said, any thoughts on which technology to go with?

I've been working as a software engineer for 9+ years. Even though graduated in CSE, I never put a great effort in learning DSA into deep. Nor did I found interest in solving problems in leetcode.

I mostly worked in startup jobs. The expectations are kind of clear here. Build apis and features and deploy.

I believe I know moderate amount of system design and build product from scratch using js tech.

Have hands on experience in Node, React and AWS. Lead ~4-5 members for a short time. Never had the opportunity to build product which needed a great scale. DB sharding, working in a system that had 100s of micrsoervices is something I have never experienced.

I'm making around $75k in UAE. How can I give myself a learning path which would lead me to a better paying job. Ideally ~120k+. Spending 2-3 hrs daily would be difficult because of personal commitment. Maybe 6-8hrs total in a week I can manage.

Thanks

Hi everyone,

So a few days ago I was assigned to design the architecture for a small feature, which I presented to the team on a meeting. During that meeting, a colleague of mine implied that he was directly involved with crafting the design which wasn’t the case indeed. How would you guys handle this? Personally, I believe just for a single case it doesn’t make it a reason to address it, but if it gets repeated it might need some careful consideration.

I'm an E5 iOS at a Big Tech company. I'm in the process of switching teams. While they're figuring out the paperwork, I've been invited to their roadmapping and architectural discussions.

The 2 E6s are dominating the discussions. The E6 iOS has been on the project since its inception a year ago and has all the context; the E6 backend has been helping out our team with the discussions and may eventually join our team. I'm the newest team member so I have the least context. The E6 backend's proposals usually sound like great ideas to me, but the E6 iOS often shoots them down and provides historical context on why those are bad ideas.

In my 1:1, my manager asked me why I'm not participating in the discussions. When I told him I don't have context, he told me to participate anyway. How do I participate effectively when I don't have context?

We are building a separate product with in a product based company most of the features are crafted just to impress the managers.

During the demo hacks are used to adjust the situation. Though I've recently realised even the API design and system design is not scalable. They are just build so that the current UX is displayed, too much hard coupling and meaningless design.

The HTTP response is dynamic as the demos were build on web and the back -end developers are not understanding that it is difficult to handle in a strictly typed languages.

I get by saying today, I'll write clean code and do my work neatly no matter what others are doing. Though I want to know how do I communicate well with them and build solid products, so that we make something re-suable and independent and sell it not just part of the current product but in terms of public APIs like Twitter, Reddit.

Lately I feel like simply discussing in these lines creates friction, but I have to speak in gentler way to get the job done effectively :)

At my startup, we use Confluence to host various documentation. I'm working on a design document draft, and I have it saved within the org's Architecture and Engineering Space (basically a folder). I have some misgivings about placing it there since I've been regularly updating it, and it notifies everyone within that group whenever I do. I'm mainly putting it there for visibility, but some of the writing is half-baked. Since it's a small company, no one really cares, but I think this might reflect poorly on me if it were a bigger company. How does Big Tech generally handle this design doc process from first to final draft? Do engineers generally just keep it within their personal Space until it's ready for review?

Hey,

I had bought a course from interview kickstart that expires on 20 days (post extension) and they are not able to extend it further.

I have 11 interviews left with them, which is really amazing given the talented interviewers they have.

I am applying for L6 (maybe L5) at Roblox, Google, Netflix and Uber.

Given the constraints, how do I best utilize the interviews:

• Is it better utilized for systems design or DSA?
• For folks who successfully clear DSA at FAANG+, how did you practice "talking" through the problems? What other resources did you use for mock interviews?

I was thinking of using ChatGPT for talking my mock interviews, is that a good idea? And using 10 / 11 for systems design and 1 for behavioral.

I was watching the system design playlist and while it's good it feels like it's frontend oriented. My question here is that let's say that we get a big epic to work on, where we have to do something similar of what is already implemented in the application, but this time we have to do something else on that similar line. The main point I want to emphasize is that we normally have screenshots from the product design side and we have to come up with models and apis and everything else(considering there are more than 10 apis to build).

What is the best way to proceed in such situations?

Can I build an android app through visual studio code or do you guys strongly recommend using android studio code instead? I was considering using VS code since that’s the platform I’m used to using and heard that it works well for android apps as well.

I have gone through a post in Slack about paper reading and its benefits.

As an Android developer, I did not find paper reading will be useful in my day to day job. Because the system design for a frontend Android is different.

I have this doubt from long time: How can i improve my design skills in Android? I see many people in our org come up with a good doubts and points in architecture discussions, but I am not able to do that. Could you please help me with what I can do to improve these design skills? I believe I should constantly do something to master this, but I am not sure what to do.

One way is I can learn it by doing - I am learning from my current project but that is not so scalable. I want to multiply my learning curve here.

Thank you in advance.

I am working on a side project. I may use GCP or AWS for it.

tldr of the system : Create a web ui to process input files and generate a tabular output giving the link to the dashboard .

I have a bunch of files generated , about 2GB in size in total , Account ID is the high level directory and inside each account there 50 log files each having specific content .

I want to create a UI using Angular framework . Using the UI I will upload the files to GCS /AWS S3 . Then the backend should trigger complex backend scripts, written will be written in Python or Java and then output file is generated. Using the output file a dashboard with few images embedded is generated.

In the end the UI should have a table which should indicate status of processing to COMPLETE and in another column show point to url of the dashboard.

I want to deal with least amount of security work as I want to avoid studying Oauth 2.0 authentication .

Few additional questions , non-system design questions =>

1. What are the things I should learn in Angular to cover most bases in this project, I am a backend engineer and going over youtube videos . I want to stick with Angular as I have few work projects going to come in coming months so it will help me to learn it .
2. I am unfamiliar with typescript . How much effort will it need to learn in order to code for this project. I am most familiar with Python only.

I am good at Systems design. It comes very naturally to me when compared to DSA in that I exhibit systems-level holistic thinking.

Could this be an indication that I should play up to my strenghts and think about becoming a Product manager or architect as opposed to struggling being a developer?

I would like to understand how can I drive an effective Tech discovery session (just had a PRD walkthrough) with the team (remotely working) before we start writing a detailed system design document. The engineers are mostly L4 and are still learning how to effectively collaborate instead of full blown meeting.

When software design tasks are given, I spend way too much time in keep looking for other wikis or resources to get information about similar problems.

I don't know when to reach out to senior engineers to get their input because I don't want them to think that I know nothing and how I even got this job. Even after some research, sometimes I can't choose an option over others despite listing pros and cons. I keep thinking about the designs all day and night till the design is finalized. How to take a break and allocate time when solving ambiguous problems?

Hello Taro!

I'm working as tech lead in company of approximately 500 people, we are creating tech products both internally and for external clients with a data-science focus.

As part of my yearly objectives, I'm working on improving technical architecture decisions across the company, and I'm looking for best practices, advice and references on the subject.

In the past we have faced some issues such as:

• tech lead looking for feedback on a target architecture not finding anyone willing/ready to do so
• an internal product whose code was so poor it had to be almost completely rewritten.
• teams struggling to solve a problem that had been already solved in other part of the company.

Actions I have been considering:

• encouraging the use of Architecture Decision Records
• set-up a team of senior technical profiles that could be mobilised to review an architecture

In general I'm weary of any solution that would:

• remove accountability from engineers
• create coupling between unrelated projects
• add a lot of meetings

Thanks a lot!

I am hitting the point in my career where I am responsible for designing newer systems to handle more novel problems. I will use my last project as an example.

I had to redesign our game engine to handle more complex scenarios. I decided to "eat the frog" and come up with a few very complex game interactions to test if any of our designs would satisfy those test cases. After about 1 month of development, we landed on a design. This design also passed all the test cases for our old engine.

I think back and wonder if I could've done it more incrementally. We didn't really ship anything during that time, just tested out different ideas in our feature branches. There are still some outstanding questions for that project that are as low level as "should we use an abstract base class or an interface for this abstraction" to "here's an even more complex interaction we don't know how to handle, but could feasibly happen".

Now I am designing a new animation pipeline and don't want to get stuck in the same 1 month long marathon with only a design doc to show for it. As a newer senior engineer and lead, a solid design doc seems like a win, but it was a tough month.

In one way, this is different than having all the work known and chunking it out. We aren't sure what we'd come up with. Maybe that's the nature of R&D. On the other hand, work is work, and I'm sure there are methodologies for making more consistent progress even in R&D.

Interested in how other people approach this dilemma. I think the crux of the issue is I wasn't sure our design would work until we built and tested it against the test cases, and I have a nagging feeling we got a little lucky.

Hello, I'm currently actively interviewing and more than once, I found myself not progressing due to some knowledge gaps on things that companies seem to expect at my level.

I worked as a backend engineer for 3 years after a coding bootcamp, mostly in product features and APIs. Distributed systems is not something I did on a daily basis.

I have struggled with system design interview questions asking about databases, concurrency, search, among other things. How can I fill those gaps to the level I can pass these interviews?

Here's what I'm going to do, and I'm looking to get advice on whether I should do, stop doing, or do something else that I didn't list.

1. Stanford Databases courses (5 total):
2. Read "Understanding Distributed Systems" by Roberto Vitillo
3. Practice system design interviews with a partner

Thank you for your help!

Recruiters often mention that a staff engineer interview involves a deeper dive compared to a senior software engineer interview, but it's not clear how much deeper it needs to go. Are there any resources that can clarify this?

Hi! I'm a new member looking to participate in the book club sessions for the Designing data intensive applications book. How can I get a copy of the book?