Side Projects Q&A and Videos

TLDR: How to pick side projects

This might have been answered multiple times and a very well-known answer would be working on side projects but I want the answer to be more in-depth so it can be helpful for many engineers.

Please don't answer in a generic way but try to answer this by posing in my shoes

Let's say you have 1.6 years of experience as a Software engineer and 3 years of experience in IT but not in development now if you want to stick to the SE career. This market is very challenging for me to get a job with 1 plus year of experience. I have to convert my IT experience into developer experience and try. But when I give interviews I tend to fail the Hiring manager rounds because they can see the depth of my SE career.

So how to convert my IT experience to SE experience? I have put a lot of effort into Leetcode and now I have gotten to a decent stage the same thing applies to System design as well I have read books blogs etc and getting the depth would be my next target.

We can hear a lot of stories in the past where a person who started his/her career as a tester or a QA and got it converted to Senior software engineer etc by working on problem-solving skills but I don't think this works in the current market.

So I felt I was missing depth. How to achieve a mid-level engineer status where I can effectively tell a lot of stories and challenges I have faced in my career and show bias for action etc

The most simple answer would be to do side projects but selecting a repository and a project is very hard as there are countless repositories and projects.

All I need is a small ignition to start on the side projects then I think discipline would take care of the rest as I was at zero questions at LeetCode a few months ago and now I have solved 250 plus with discipline.

I want to be able to work at FAANGMULA and am considered an industry leader publishing work in my domain, however I am told the culture of Apple, Meta, Google (all of whom which called to hire me) is not open to this (this is all lower level employees whereas most of my friends who are directors of engineering all say I can negotiate this for my next big work and that I can just disclose this an interview and should be fine). Full disclosure: I've never worked at a larger FAANGMULA company (have either founded startups and worked on small teams, consulting etc.).

I am wondering if it is better to place my autonomy over a bigger salary and stability at a larger tech company vs an earlier stage AI startup (maybe not all the perks in the world), but is friendlier to my contributions outside of work (it is directly related) where they don't care and probably encourage that I do publish. How should I think about this in terms of long-term career growth?

I think that there is a chance the startup fails to productionize faster than I would prefer (they're a market leader already at the very top of FAANGMULA developing new foundation models are some of the OG AI people best in class), I wonder if the path I take is better at a FAANGMULA company who can likely compensate for further graduate studies or cover other benefits/perks and higher salary than at the startup, however, it would mean giving up being able to publish and feeling a sense of autonomy.

I'd love to hear any decisions people made building side projects that you learned the hard way was a bad choice

I'll start:

Deploying ML models for a project is a pain. ML models are often 100-200 MB so you cant just deploy on a lightweight VM and it exceeds free tiers. I didn't want to pay to deploy a side project that has no users.

I tried digital ocean (i got a free 300$ credit) and it was a PAIN. I spent days 😭getting it working to no success. Eventually I used and it worked really well. Highly recommend. I wish I didn't try and deploy on a VM or use digital ocean for ML. Railway worked because it seamlessly downloads all the libraries from the code like heroku but cheaper but digital ocean's version of this had so many limitations with memory and ram and storage

So I'm looking at exploring contractor roles more seriously. I've been a contractor before but I fell into those roles. I'm trying to be more proactive in finding contractor roles but I wanted to ask for advice as I start to work on this.

What were the most useful things you did to find contractor roles?

I'm not too happy with my current job, I feel like career growth prospects are dim and the work is not challenging enough. Given this, my best career move is to find a job at a different company.

Given that the work at my job is not good, I want to use side projects (which Taro tells me has a ton of benefits) to be able to get "equivalent" job experience so that when I interview at other companies I'm able to leverage the experience with awesome side projects to do well.

More specifically most data engineering jobs want experience with airflow and pyspark. My current job only has me doing SQL stuff, so I want a side project that leverages airflow/spark

Problem: If I focus on just building apps/web apps to get users then that wont help my data engineering career because most apps dont need someone with a full pyspark pipeline with airflow and whatnot

Conversely, if I focus on building an awsome data engineering pipeline, I'm likely not solving any real world issues and have 0 users, but the skills will help.

Problem 2: If I focus on consumer apps I'll have to learn react (which I 100% dont use at work) and spend time doing backend stuff which isnt helping me grow personally as a DE because I read on Taro that you should be going deep, not wide.

Rahul hit the nail on the head on the recipe finder app I was thinking about.

• The biggest blocker for is that I just don't care enough (or I'm too lazy) to try out new recipes. I just default to things I've made before, a ready-made meal, or Chipotle.

The YC founders talk about how "discovery" (music/band/concert/event/apps/food/restaurant) apps a tarpit idea -- everyone thinks of these ideas, realizes its fundamentally broken, but no one can fix it

The reason -- people default to the most common option or things that are popular. And I noticed that with this app as well

My new suggestion is a recipe tracker instead. Paste a link and it organizes/sorts it

Pros

• Even simpler (is an "L1" idea - utility tool/no need for dataset)
• Scouted competition. There's 5-7 apps on the play store that do this with 10k-1M downloads and >4.5 stars. So there is demand.
• All these apps seem to have a freemium model. Either ads or save only 20 recipes which is of course annoying. Taking a page out of Alex's playbook this is a good opportunity

Cons

• I'm not looking to do this as an android/ios project because android/ios is so far removed from typical ML/backend ML. So I'll want to do this as a web app
• Concerned getting traction on this as a web app. I'm not sure how to promote it or get people to use it. I don't know anyone who would actually want to use and not use it because I'm asking/making

Questions:

1. Is learning design a worthwhile investment as someone most interested in doing full-stack work at product-based startups?

2. Might working in a small product-based startup be an effective way to pick up design skills while working as a SWE?

3. How can engineers build more complex side projects without any design skills?

Regarding #3:

I’d argue that basic product design skills are critical for building any CRUD application. You can’t build something without defining what it’s going to do first. 

Literally - you can’t write code for a feature if you don’t know how the app will behave during a loading state, an error state, a complex edge case from a wonky user flow, etc.

You can wing the design and iteratively dogfood it to improve its UX - but that’s the same as doing UX design while having zero UX design skills. It’s the software engineering equivalent of writing spaghetti code - except you’re not even improving.

Personally, I find that UI libraries like or are most helpful for solving UI problems like designing a button or a modal. However, they can’t help you decide how a screen in an app should work, nor can they abstract away all design challenges for more custom use cases.

Also, any CRUD application built with poor design will inevitably feel like a crappy database client. 

The design problem applies to backend projects, too. Backends exist to service frontends, so you can’t build a backend without knowing what features the frontend needs - and you can’t do that, either, if you don’t design it first!

These are all challenges I’ve faced working on my own projects.

I suspect the best approach is really to just learn UX design and a design tool like Figma. However, that’d be a hefty investment given UX design is a separate field from SWE - especially if it’s just for a side project.

Also, building cool stuff as a semi-competent engineer is tons more fun (for me) than learning design from scratch!

What are your thoughts on my aforementioned questions?

I'm looking to deploy my project to get some users. I wanted to ask what the best way services to use as I'm only trying to show off the work mainly and maybe get some users once I get unity app on the stores.

Why does Disney Plus use an entirely different video component when wanting to display ads?

During my making of a browser extension that either manipulates current subtitles or adds new ones to a video on various platforms:

I have to partially reverse engineer how each site is going about the displaying of subtitles and the handling of their video client.

I found myself encountering a random edge case with Disney+ that every 5 - 10 reloads of a video - my extension would break due to the expected elements not being there.

What was happening was that Disney+ would add a completely different video client with completely overwritten HTML5 video element APIs and an arbitrary offset of 20 seconds to the time.

Other video platforms such as YouTube, Netflix etc don't function like this and have a standard video client with adverts working with that standard client.

Is this just bad engineering from Disney Plus's part or part of some type of AB test or is there something more technical I haven't thought about?

Tricky to answer for certain unless you've worked there but I'm interested in the different perspectives.

I get that getting users means the project is impressive which makes it a better project which would lead to more interviews.

But why are projects with users better?

Isn't a project with users just giving me signal about your marketing/sales/product skills and not engineering skills? Especially if you're building a utility app/one time use app e.g. background remover. The hard part about getting users isnt the technical stuff its the marketing/sales based on the fact that most apps I've seen ship something small, market it a ton, get users, and then worry about making the app really really good.

Would using no code tools to build a site which got a 1000 users be just as impressive if I used nextjs and nodejs?

Recently, Rahul has in terms of their ability to deliver business outcomes quickly and with minimal learning. They do look cool!

As a data engineer with some backend experience and no frontend experience, when I think about making a web-app side-project, what should I do first: use a no-code platform to build it, or learn the basics of HTML, CSS, and Javascript (along with a framework like React)? Should my default be to use a low/no code tool?

More broadly, how should I integrate these into side-projects and my work in general?

Thanks!

Scenario 1: If I share personally identifiable information about myself in great detail, could another user access that or how is my privacy protected?

Scenario 2: If I'm working on a project that could compete with a Microsoft or OpenAi ML project, and I put a data set I’m using into chatgpt and share details about what/how I’m doing stuff, do you think they would bother with flagging something if it was potentially useful for themselves? Asking for a friend.

My goal is to drive traffic to my app by making youtube shorts. Specifically there are interesting facts about compound interest that could be genuinely interesting for people thinking about investing or taking out a loan. The idea is the graphics will be screenshots/video from my app.

What tool do you all use for video overlay screen recording like this? @Rahul you make super bright and colorful graphics for your youtube cover graphics plus in the videos - is this from photoshop or something else? Also you have a lot of stock graphics and video footage - would you mind sharing if that is paid or free?

Thanks!

I admittedly got lazy with not creating a design doc for my side project cuz I was wanting to make something for myself.

I’m now keeping one to document what I want to do next with it and scope out all the misses. Is this usually how side projects end up going like?

Been working on anything cool lately?
Need help with a bug you've been encountering?
Want to share a sick demo?

Post away!

I built a web app for a side project but its hard to SEO optimize it because there's not many words. It's mostly a utility tool project e.g. fitness tracker. Only blogs seem to rank really high on websites