Looking for career switching advice from Film/Business to Software Engineering
I have a few weeks left before I finish my MBA and I turn 26. I decided to spend the free time I have since my cancelled finance internship learning how to program. I enjoy it a lot and could easily see myself working as a Software Engineer. Prior to my MBA and the finance internships I did, I was a video editor and colorist with a film production bachelors.
I started with App Academy's Introduction to Ruby I and II courses but I quickly realized that even if a boot camp would work out for me, that specific one would not work out due to geographic location. There is a Hack Reactor branch close to my parents house in Los Angeles, however I feel like if I choose to enroll in a boot camp I should learn as much as I can on my own first; doing it now seems like I would be subjecting myself to financial peril, especially in light of COVID-19.
I am now taking the edX MIT 6.00.1 Introduction to Computer Science and Programming Using Python course. One of my friends gave me access to a udemy MTA Networking Fundamentals course and I was considering working through that alongside the MIT one. The MIT course only takes around 15 hours per week, so I need more classes/projects to help fill what is ideally a 50 hour work week in order for me to really grasp these concepts. I know at some point I have to begin to build projects, but I do not believe my fundamentals are quite there. If I'm wrong, I would love some pointers as to how to begin.
If anyone is able, I would really appreciate some guidance in three main areas. First, is my timetable realistic? I have enough overhead and family help to last me to December before I start looking for work OR enroll in a boot camp, the latter of which would require some additional help. I feel that if I diligently study from now until then my family would be willing to help me out a little bit.
Second, would a boot camp actually be worth it? I realize that a solid portfolio is the #1 thing that will get my foot in the door somewhere but I'm not sure if I would have one that is as fleshed out as I might need by December. Some alumni from my school went through Hack Reactor, but the amount of CS students currently working as SWE far outnumbers the two folks I have found so far that pivoted from Economics and Film Production into SWE. I see the boot camp as serving two purposes: demonstrating that I have what it takes to commit to something and see it through (similar to my two degrees, just for programming) and helping add a few items to my body of work. The financial obligation is a bit concerning, but if I truly was able to find a job at the end of it the ISA would feel worth it.
Finally, how can my lesson plan be improved? I intend to finish this MITx course over the next six weeks (I'm working through week 2 right now). Is the MTA Networking Fundamentals course a good compliment? Are there other courses that might be better? I own the book Programming Principles and Practice using C++ by Bjarne Stroustrup and have read that it is a solid introduction to both C++ and programming fundamentals, similar to the MITx course. I intend on starting and finishing a Version Control with Git coursera class this coming week. What more could I add, are these the right things to focus on, and is this even a realistic endeavor for me?
My interests are web development, app development, and data analytics. I feel each of these would be its own path, but if someone could help me narrow this list down to one that best suits my past experience, has complimentary skills, and seems like it would be a good fit it would be much appreciated.
Any advice helps!
First, just be reasonable in expectations. There are a billion stories about self-taught people working at big tech companies; those are unicorns. It's not the norm. In fact, the only reason why SWE has become such a hotspot for these, is less regulation. There's no reason you can't self-study to become a mechanical engineer in theory either, it's just that the government won't let you call yourself an engineer unless you go to an accredited bachelors program.
Missing out the degree is a big deal. Not only will you be below degree holders in priority, but you're missing out on an entire pipeline for junior developers that many companies have built up and trust. That's where going to "prestigious" universities plays a part--it's not necessarily that they reject everyone who comes from Podunk, OH, it's that the pipeline. And also the networking.
So I'd just say first of all, expect an uphill battle. The economy is not great, just about everywhere in the world, there's honestly quite a bit of CS newgrads looking for work right now.
In terms of courses, you won't get the fancy virtual diploma at the end, but many large universities don't wall off their course material. There's nothing stopping you from going through them, lectures and all.
Here's Berkeley's intro course, data structures for instance. I would also consider Algorithms since it's critical for interviews
Maybe test yourself with an old exam
Speaking of that, unfortunate, but start doing leetcode. It's spread everywhere, from "prestigious" roles to random small companies. Leetcode style problems is how interviews are done, and you'd want to at least be able to do LC Med problems by yourself in under an hour.
I absolutely do not expect do end up in a big tech company any time soon (if ever) but I do intend to practice the things that they look for in a candidate, assuming those are best practices.
I looked into masters programs and Georgia Tech's online CS masters seems very highly ranked and is well within my budget. If I land a job in SWE, I intend to do everything in my power to earn a CS masters from this program.
I agree that it is an uphill battle, but I believe that nothing worth doing comes easily. I will make this work.
Thank you very much for pointing me in the direction of the course material for Berkeley; anything to help me grasp the material is much appreciated. I also intend to work through whatever CS courses I can find on OCW.
I signed up for leetcode yesterday and plan on working through problems on a regular basis. I actually cut video games from my life completely a few weeks ago and from what I gathered, these kinds of problems fill a similar niche as the kinds of games I used to play.
Just be aware that LeetCode and similar sites have some issues. Their questions are often poorly written and hard to understand, and when you're doing the problems and trying to figure them out, you often aren't given the inputs that are failing (unless it's changed since I last tried it?). They can be fun, but they're not necessarily a realistic scenario like you'd have when writing code on the job. You can't ask questions to clarify what something means, and they often focus on performance while completely ignoring readability and maintainability, whereas in a job, the opposite is often the case.
I'm pretty sure you're talking about Hackerrank instead, but regardless neither is really for actually getting better at software engineering. Rather, it's because they're the same format as whiteboard interviews, which use to be mainly used by major firms, but now are basically everywhere.
These days, if you want a SWE job, you'll probably have to do a leetcode style whiteboard problem.
Thanks for the heads up! I've read that a lot of companies use LeetCode as a screening tool. I enjoy these kinds of problems so working on them from time to time seems like fun, even if it doesn't fully encapsulate best practices.
I made a switch from publishing (was a liberal arts major) to software engineering.
I was self taught, and so I can say it definitely can be done. It's hard to comment on timetables etc, there are so many variables in play. FWIW 6 months seems a bit tight. If it's a hard deadline, then so be it and you gotta work within that constraint. Either way, after 6 months of consistent self study (or some other route like a bootcamp), you will have learned a ton and you will be that much closer to your career switch. At minimum you will have gained a very valuable and useful skill to go along with your MBA. I wouldn't discount that at all.
For me, I taught myself over the course of two years while I had a full time job. I intentionally avoided setting a deadline for myself because I wanted to avoid burning out from the pressure of meeting that deadline. This of course is a luxury i had, since I had a good low stress job and could afford to take my time. But I would emphasize that if you genuinely enjoy programming, you will find yourself propelled much further and more easily when you are self teaching if you are learning material, building projects, reading for learnings sake (ie, not with the explicitly conscious intention of learning X for your resume). When self teaching, I found that things worked better for me if I focused more on the enjoyable pursuit of learning software as a hobby, and that if I followed this approach, one day I would wake up and know enough to get a job. When I found myself in a stretch where I was learning or doing something purely for the resume, I got much more fatigued.
FWIW, I did about 1.5 years of python and then 6 months of Ruby (some sql and JavaScript thrown in there too). I got a job at a rails shop (Ruby), though i do some python too. My advice is always to follow your interests, though, so not making an arg for those languages over some of the others you mentioned (eg c++).
As for learning fundamentals vs projects etc, i think this is a common but unhelpful framing. Programming is all about learning a variety of tools and putting them to use. You don't need to be a seamstress to sew a button on a shirt. In my former life, I was a copyeditor and I noticed that a frequent mistake folks would make would be to accidentally duplicate some word -- "I went to the the store". So one of my first projects was called "check4repeats" that scanned a text file for duplicate words. I also made a command line hangman game. These projects were bad! But getting into the habit of "okay I learned a few new things (tools), what small thing could I make to give them a spin?" will break you out of the tutorial, tutorial, tutorial cycle and allow you to learn to figure stuff out with fewer guardrails. It will also help you find your way to tools and concepts you didn't know you need to know ("ok how do I make this button do X? How do I store the users score in a database?")
Wrt bootcamps, I should mention, I've worked with bootcampers as well, and my company recently participated in a bootcamp show/tell, where every candidate there got hired. So from my perspective (probably depending on the bootcamp) bootcamps are absolutely a viable route. Obviously in most cases you have to put down money, can't keep a job, the time/day commitment is much greater than it might be if you're self taught. But they seem to work for many people. A lot of it probably comes down to style -- do you prefer structured guidance, a la a classroom vs unstructured self pursuit.
Finally, I should mention: You have an MBA, that should help you out. I was in publishing and got a job as a SWE at a publisher. Professional SWE is all about leveraging your programming skills and knowledge to solve business (or other kinds of) problems. So don't neglect that angle or think that your previous experience is not applicable.
Thank you for this. I haven't found many folks that have completely shifted gears so its nice to hear from someone who has been in my position. You might be right that 6 months is a tight window, especially with how things are right now, to fully realize my goal but I think at a minimum I can leverage what I've learned into a role where I can continue to grow as a SWE, even if minimally. Progress is progress.
Your insight about how I can use programming more practically is very helpful - I'll see where I can apply it to the things I already do or what friends and family might need help with to start. Maybe simple scripts that just help day to day stuff.
That's nice to hear about boot camps, actually. I think I will revisit the option after all come December. I actually thought of a tool I could create that might simplify things at a previous internship, so once I have a bit more of the syntax of Python down I plan on starting with that. It's a simple inventory management system that I'm sure has been created a million times but I think it would be a fun project to take on.
Inventory management app sounds like a great project!
I'm a film major trying to break into computer science. I have this advice:
Thank you very much for this.
I have not cut ties with anything, actually, and I intend on doing some editing and color work in the near future to keep the bills paid.
I started taking detailed notes for the 6.00.1 class but you're right, I'm not being as efficient as I can.
Thank you so much again for this advice. I am fully prepared to fail and fail often but that's the only way to move forward.
These days it seems like VS Code is becoming more and more popular. Why do you recommend the older editors to beginners?
I do not. The second part of the list item was kind of a joke. There are many advantages in using editors that were created a long time ago but remain current, useful, and relevant (small rant). But I won’t lead anyone down that path, the benefits are minimal unless you have certain needs and predispositions.
I mean VSCode doesn't fulfill all niches. Oftentimes it's way too unwieldy to use, and just a terminal text editor is worth learning for when you need to make quick edits, or to very large log files.
Not only that, but Vim and to a lesser extent Emacs are more than just text editors, they're modes of control for editing. For instance, I use vim keybindings in VSCode, and vim keybindings are also in most common IDEs at this point.
Yeah, a lot of people seem to like vi key bindings enough to make them work in other editors, which seems like a point in favor of learning them. I only know enough vi to make a quick edit and quit, but that's useful to know.
For work, the project I'm on uses a (mostly) Scala backend and a (mostly) React frontend. I use VS Code for the React, and Intellij for Scala. I don't think it's possible to use a java-based language without an IDE and remain sane. In years past, I've used a relatively bare-bones vim setup to write PHP, Perl, Go, and Node.js.
I still use vim for quick edits and miscellaneous text. Just need to update a config value? I can open, edit, and save with vim faster than the splash screen loading up Intellij. Fiddling with an SQL query while looking at
psqloutput and maybe some other system command outputs?vimis sufficient and lets me keep it in a terminal split-pane with the rest.If nothing else, one should be at least familiar with the basics of
vi(m)even if it's not their daily-driver because it's practically everywhere. If you have to ssh to a remote server to edit a file, a fancier editor may have a remote option somewhere, but vim's probably there. If you have to use someone else's machine and they don't have your editor of choice installed, vim's probably there (unless they're developing on windows, but, really, who does that?).You seem pretty smart and motivated, so I think you could get enough experience in before December to get a job in software engineering, likely an internship (although software engineering internships often pay as well or better than normal entry level positions for other fields).
In my only partial-qualified opinion, bootcamps and the general idea that people can go from 0 to junior software engineer in just a few months is delusional. There's a reason people spend 4+ years in college learning both computer science principles and software engineering skills. If bootcamps did even 75% of a college education no one in SWE would waste their money on college.
That all said - smart students with only 6 months of programming experience will easily land internship positions. If you're okay spending 12-18 months as an intern you could make your way to a junior developer position 2 years from now.
If you have time you should read books on software engineering in addition to programming. The Pragmatic Programmer is one I highly recommend. Clean Code is one I haven't read but have heard many good things about. But I wouldn't worry about these kinds of books until you have some experience programming - else you may have no experiences to relate the advice to.
I would be thrilled to land an internship and I'm glad to hear that it could be within my grasp by December.
I agree about boot camps with one caveat. I imagine for certain roles, not all of the knowledge one gains from an undergrad/graduate degree are utilized. Many undergrad degrees also have a general education component that takes roughly 2 years to complete. I would imagine if a boot camp grad can even approach 50% of the skills that a CS undergrad and they have practical work and life experience that they might be solid candidates for some roles. Again, it is purely anecdotal, but two boot camp grads from my school have full time SWE careers, one of whom is at Amazon as of last year. It seems that it can work for some individuals and it might help supplement my portfolio. I think I'll revisit this question closer to the end of the year. I likely will not enroll in one, as the Georgia Tech CS Masters seems like it is a much more fulfilling and comprehensive education with a more realistic time horizon at a fraction of the cost.
I mentioned in the post above that I cut video games out of my life entirely. A lot of my free time growing up was spent playing them, but I've found since I got married that I have been enjoying them less and less and so finally pulled the plug completely after a brief stint with PSO2. I have a lot of time to read again and will definitely read The Pragmatic Programmer. I might read them sooner rather than later, as they might help shape my programming skills similar to how earning my MBA at what I would consider an early age (much sooner than it has relevance, anyway) has shaped my personal development. Plus, I can always revisit them later on! If you have any more recommendations, I'm more than willing to check them out!
I'd just note that the take above is imo waaay too optimistic re: internships. Not only are many of them explicitly only for bachelors students, but a lot of them were canceled.
Here's a long ass list: https://github.com/gcreddy42/hiring2020#covid-19-summer-2020-internship-status-list
Ah, I’d totally not considered the state of internships in the pandemic world.
Wow that's a sobering list. I'll definitely keep it in mind as I move forward. Thank you!
If bootcamps were 2 year programs then I agree they'd be comparable to the CS/SE education you can get from an undergrad education.
If you put anything up on GitHub feel free to send it my way and I'll give you a full code review (unless it's in some super esoteric language).
Fair!
Thank you very much for your help and I will definitely let you know when I have things on GitHub!
Yeah. I rarely watch film and TV these days and my interests in life have little overlap with the entertainment industry. I live a very utilitarian/minimalist lifestyle and can fit everything I own (outside of furniture, some kitchen stuff, and a few large appliances) into my computer desk. I care a lot about minimizing resource use and maximizing efficiency when it comes to everyday things. I would like to do work in life that directly contributes toward a better world rather than creating simulacrums.
My MBA internships were in the restaurant industry and the film industry. My experience thus far lends itself to going deeper into entertainment on the FP&A/Production Accounting side, but I really want to move out of Los Angeles some day and I feel that I'll be stuck here if I keep going down that road.
I'm happy to pivot more gradually, but that seems predicated on my landing a job more in the business side of things than the film side. I have some color grading work coming up soon so I'm not hurting yet for cash but I am losing health insurance soon so ideally I commit to a pivot now that will better set me up for the future than be tied down to something due to benefits and past experience.
I've used DaVinci Resolve, Premiere Pro, After Effects, and Maya. All of them have a plugin ecosystem. I have little interest in plugin development for these.
I am very interested in image/video processing, thanks for these recommendations!
I discovered 'automate the boring stuff' a few weeks ago and I intend to go through it at some point since those seem like generally useful skills. Especially the Excel stuff.
I'd actually love to hear your thoughts on media if you don't mind as I have many of my own!
It's tough but absolutely manageable. I have an amazing support system and I think now is the best chance I have to pivot. My wife and I are currently staying on her family's ranch, which bizarrely has better internet than I had in LA so I have every tool I need to make the switch.
That's great to hear and I will absolutely begin contributing to open source projects as soon as I can. With my color work I often volunteered time to anyone that needed help since the difference between a professional and amateur is often in the post production (assuming the cinematographer did a decent job). Are there any you know about that a beginner could assist?
Another way you could transition into software is through product management. I'm biased because it's my field, but is viable as a recent MBA grad. You'll be able to leverage your degree into a position where you work closely with developers. You'll find opportunities to write software to help the team. Then you can figure out where to go after that.
My favorite thought leader in product management published this the other day. If you want to know more, let me know.
https://svpg.com/product-management-start-here/
I hadn't actually looked into this and it's really interesting. I would like to know more and it sounds like it could be a great way to get to where I want.
Thank you for sharing, I fully agree! For me where it gets fuzzy is the commercialization of media to fund raise the more risky projects. Also, I feel that there are diminishing returns on quality - especially when an idea is created, then copied, then copied over and over until the original message is lost. It's super interesting and definitely something I would like to spend more time analyzing about even if I don't directly take part in the conversation (for now, anyway). In short it's why I like tildes so much - it seems to be taking a bottom up approach that tackles what I find to be the darker side of media production.
That's alright, I think ultimately I'll discover what sorts of projects I would like to actively contribute to organically. For now, I actually hardly use anything outside of the basics. On my iPhone I have 5 3rd party apps: AdGuard Pro, 1password, my banking app, Venmo, and LINE so I can keep in contact with one of my friends abroad. On my laptop (iOS) and desktop (Windows) I have Zoom, Dropbox, Microsoft Office, DaVinci Resolve, Photoshop, Lightroom, Malwarebytes, emacs (haven't used it much yet), and Anaconda for Spyder since that's what the 6.00.1 course recommended I install.
Websites I visit are Google, YouTube, edX, MIT OCW, Tildes, Reddit, GitHub (soon), LinkedIn, Gmail, Protonmail, Google Docs, Amazon, Netflix, chess.com, online-go, Roll20, Myth-Weavers, LeetCode (soon), government sites, and whatever else google sends my way when I'm looking for answers. It feels like my internet use is pretty bare bones, especially since the extent to which I use social media is limited to LinkedIn, Tildes, and Reddit.
I guess by default I would be interested in helping contribute to any of those since those are actually the things I use.
Regarding this - I highly recommend the Stack Overflow Code Review site. You can get some really decent feedback on your projects there. Just be sure to read over the requirements for posting there, as they're a little non-obvious.
Just bookmarked this, thanks!
To elaborate on what @radiator said - I work on the software side of film and video editing and motion graphics. If you are interested in that area I would be happy to share my experiences in the field. I've done work with both Adobe and Apple on most of their film/video products, and sold my own plug-ins for them for quite a while in the past.
I wrote in my response to @radiator that I do not have a lot of interest in developing plugins, but perhaps I do not know enough to really say that with as much confidence as I thought. Would stuff like Light Right or Go Figure fall into this category? I know of things like Magic Bullet and Trapcode that have sped up my workflow considerably, but as I wrote below I don't really want to stay in film world.
I'm not familiar with those products. Do you have a link? Their names are so generic that it brings up a bunch of unrelated things when I search for it.
But in general, there are a number of types of plug-ins available for most video editing software. I mainly deal with image processing plugins. These include things like color correction and color styling like you probably did as a colorist. They can also be distortions, blurs, glows, compositing/keying, format conversion, etc. There are also plug-ins that render 2D or 3D stuff from scratch.
In addition to image processing there are workflow plug-ins and apps. These generally process video editing project data, such as edit decision lists (EDL), color decision lists (CDL), or XML projects from the apps you use. They have a variety of uses, but are generally used to make the mundane parts of editing easier. (This is in the category of "automate the boring stuff" mentioned above. See for example Producer's Best Friend.)
No big deal if it doesn't interest you, but if you have any questions about it, feel free to ask.
It interests me generally but not so much career wise. I will definitely ask questions that come up!
Light Right
https://www.youtube.com/watch?v=3826cFEnWd0
Go Figure
https://www.youtube.com/watch?v=xEdZOifROmk
Ah, neat. Yeah, those are definitely the sorts of things I'm talking about. Very cool!
I will likely have some questions regarding image processing in the future so thank you very much for your offer!
I would be wary of boot camps but even more wary of spending a lot of money on school tuition and ending up with a lot of student loans for a substandard education, and there's a lot of risk of that due to coronavirus. All the traditional schools are figuring it out as they go and I think a lot of students will be unhappy with the results.
It seems like a good time to give inexpensive remote learning a try? There might be a bootcamp that is already remote-only (so they've worked out the bugs) and is also good? Also, the companies hiring people out of bootcamp would likely have somewhat lower expectations at first, expecting you to learn more on the job.
Unfortunately I don't know what to recommend. I'd look at Lambda School since they are online-only, but they got some bad press and you should probably read that first.
I don't think of getting a job without a computer science degree as being a "unicorn" thing since I've worked with people who had other degrees. Some people get in sideways through a job in QA or support, and then end up programming.
I just found MIT's computer science degree requirements
http://catalog.mit.edu/degree-charts/computer-science-engineering-course-6-3/
and the courses themselves!
https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/
After having spent six of the last eight years in school, I feel that more often than not my classes were essentially glorified reading lists. Class time helped me commit the content to memory and further understand the concepts, but attempting to apply the knowledge to projects and the real world seems like it helps me learn the fastest. A lot of my work these past few years has been color grading - something I never took a course for. I think the above MIT courses and associated work might be the perfect thing for me these next 6 months.
I feel a lot of schools might readjust their pricing models given the availability of stellar online courses and cheap degrees (Georgia Tech's Online CS Masters costs 6.6k TOTAL).
Yeah, that does seem like a good deal.
By the way, if you're looking for an interesting project to try, how about fixing some minor bug you care about in Tildes? This will give you a taste of how programming works when you're working on real software, starting with how to get it running locally and understanding the code.
(I do feel guilty about not contributing patches to Tildes myself. In my retirement it seems I've gotten lazy about programming.)
IMO you contribute in plenty of other valuable ways, @skybrian, e.g. all your submissions + usually well reasoned and/or informational comments... so you don't really need to feel guilty about not contributing code as well. :)
Oh I know. Thanks!