Has anyone here graduated from this program? I'm curious about the instruction quality and career prospects in comparison to in-person programs from similarly prestigious schools.
Not graduated yet, I'm about at 40%. So far it's very interesting and well done from an educational point of view. There're some rough points still - you can see the classes were originally taught at the brick-and-mortar university and they have been "ported" to online, plus sometimes there're some glitches between different systems being employed, but the overall quality is astonishing, especially for the price.
Probably an in-person MSc will be a bit better than OMSCS, but it will cost 7x-8x and would require people to move around the country.
That's about my experience too, having done only two classes so far. One of the largest downsides (in my experience) is never meeting/hearing from the instructors of the courses- the TAs are almost solely responsible for running the classes. I went to a large state school for undergrad, so lack of instructor facetime has not really been an issue for me. I've also found many of the udacity lectures to be so-so, but the GA tech projects really help to drill information into your brain and can be quite time-consuming.
Yeah, people who think online ==== easy will not have a good time. I love OMSCS, I'm about to start HCI this summer. It's great because I have a mortgage and a career - the amount of debt to go full time would be staggering, and the cost of attending private classes would ruin my long term financial goals. Also, I just really like it.
I'm on my 9th class, one more to go. I found that Machine Learning had decent involvement from Isbell (he'd chat with us on Slack) and that Operating Systems and HPC and excellent professor participation -- they would hold weekly office hours (video conferenced) and were active on the forums.
There were 1 or 2 of my earlier classes that did feel phoned in, though (but did have strong TA involvement at least).
I graduated from the program in December. Overall, the program was absolutely worth it for me. Based on my experience I'd say that the online program isn't quite at the same level of instruction quality as an equivalent in-person program simply due to the number of students in each class. However, for me personally the lack of cost (both real cost and opportunity cost from lost work years when pursuing a fulltime MS) more than made up for it.
I haven't interviewed or applied for jobs since graduating, but it's worth noting that the degree itself is the same as the one awarded in-person. If you get involved in the community, the network is also surprisingly good; the slack channel has hundreds of students at all sorts of places.
I was admitted to the program in the pilot cohort in 2014 and graduated in 2016. I left my civil service job as a mechanical engineer with the DoD 3 months later to take a job in SV, and I just started a new job two weeks ago at a FAANG company as an ML Engineer. I think having an MSCS from GT helped me get here.
I don’t know if my experience is considered typical. I worked as a TA in the algorithms course and then in Computational Photography while I was there. The algorithms professor was actually how I found the SV job I moved here for. I’ve seen it mentioned elsewhere already, but I can confirm that TAs run most classes—but it’s also worth pointing out that TAs in OMSCS are often comically overqualified, and 90% of the work in running a course is logistics (due dates, extensions, grading, etc.).
I think it’s as valuable as a resident program if you can’t justify leaving your current job or taking on debt to finance an advanced degree. Just make sure you try to learn everything you can while you’re in it.
The program did not initially allow remote TAs, but they started allowing it after a year or two. The latest data I’ve seen is that about 50-60% of the TAs are also OMSCS students and the rest are from the resident program in Atlanta. It was my experience that GT professors preferred to have a senior TA or head TA from campus (usually one of their PhD students), but they were comfortable letting most of the graders be remote. Some online TAs earn a reputation for being awesome and wind up as head TAs, and some profs don’t mind having fully remote teams, etc., but that’s the gist of it.
I’m about halfway through and also did my undergrad at GT. Overall I’ve been pretty disappointed. The lectures are just YouTube videos played through Udacity, and the quality has varied from okay to bad. TAs make a big difference, but the quality there is a huge range. Some classes have had really good interesting active Slack communities, while others are basically just a Piazza page with people asking for help on projects. If you were disciplined and didn’t care about the degree at the end, you’d get a lot more bang for your buck just by watching the YouTube videos and reading the textbook, since that’s basically the level of instruction you get.
> If you were disciplined and didn’t care about the degree at the end, you’d get a lot more bang for your buck just by watching the YouTube videos and reading the textbook, since that’s basically the level of instruction you get.
Comparing it to my Master’s which has even less faculty involvement I doubt it. Watching videos and reading the textbook leads to the feeling you know something. Doing projects and preparing for exams, that leads to genuine learning. You learn far more working through past exam papers or doing all the questions in the textbook than you possibly could by trying to absorb information passively.
You would not get any of the class interaction, you would not get any of the projects, and you would not get any of the feedback from the projects. You would miss a ton of knowledge by missing those.
Also, the slack community makes the entire program -- you really learn a lot watching the people who know the material help people out through the projects.
I graduated 2 years ago next month and will say that doing the OMSCS was the best professional decision I've made. There's one unusual thing about my case though: I have a lot of prestige on my resume (Ivy league school, BCG, Google), so I mostly just wanted a) the learning and b) to say I have an MS. I don't know how I would feel about the program if I were more concerned about prestige.
Another 2017 OMSCS grad with a unique career path chiming in. I was a MechE by background, but had some pre-existing computer science education from dabbling with robotics. My primary goal was to broaden my skillset and achieve the M.S. designation. To be honest, the M.S. has been nearly useless (nobody asks/cares), but I'd do the program again in a heartbeat, as I'm now a VC that is at least somewhat conversant in many of the technologies we are encountering.
I'll be roughly half-way through at the end of the semester.
It's the hardest thing I've ever done, and well worth it.
The quality of instruction varies wildly. OMSCentral is a student-run review site that covers the courses well.
The logistics depend almost entirely on the quality of the TAs in the courses -- there are courses like AI4R where the professor is heavily involved and answers all student questions, but the majority I've seen the professors are very hands-off and the TAs handle all student interaction.
Is this your first semester? The waitlists clear up shockingly aggressively when the semester starts. I'm always surprised when I get into a class from position 300 on the waitlist, but it always happens.
The thread is filled with other OMSCS students, so I may as well throw in my perspective as a grad.
- The instruction quality is comparable to free online courses (which is unsurprising since it's all delivered over Udacity anyway.)
- I learned a ton
- What Georgia Tech gets right is simple and a little sad: credentials matter. I learned a lot by going through a bunch of online courses and being graded, but a significant chunk of the value is just saying I have the degree. I wish that "I took a dozen MOOCs" carried the same weight as saying that you have an MS from a top-10 school, since in this case they really are exactly the same, but it doesn't. That's particularly true with arms-length interactions. Day-to-day my CEO cares more about the skills I picked up from the program, but when he's assembling slides about team for a pitch deck, the credential matters more.
That's a good point. I haven't really come across bad online content, but that's probably because I've stuck to stuff coming out of top universities rather than random Udemy courses.
SDE in AWS working in industry for approximately 10 years.
I'm finishing up my first semester, taking graduate introduction to operating systems (GIOS), and so far I'm super impressed with quality of the program (again, not sure how the other courses compare but I hear GIOS sets the bar quite high). In addition to the rigorous curriculum, the students actively chat with one another online; the current course has a very active Slack instance — paid out of pocket by one of the TA — where people chit chat pretty much every day, many of us sharing our frustrations with the projects that, as others have mentioned, drill the theoretical concepts into your head.
In short, I'm pleased with the program so far and look forward to the rest of the courses.
I graduated in December. Before this I already had an MS in Finance.
Overall OMSCS was a phenomenal value. The lectures are hit or miss, but that's not where the value lies in general, though there are exceptions (Joyner's classes like Human Computer Interaction). The value of the classes lies in the curated projects that high class professors designed and selected to facilitate learning. I learned so much in RL and ML especially due to those projects, as well as Educational Technology due to a project I selected myself.
FWIW I was working full time as a new software engineer when I started the program and after a year it helped me move internally into a data scientist position where I've now been for 2+ years.
My path was BA philosophy -> MS Finance (worked as an investment advisor) -> after a year break I went back for a post-bacc certificate in CS during which I interned as a software engineer -> OMSCS, which I started the same month I started my first full time software engineer position. A year later I transitioned to a junior DS position, and am now a DS. I'm 30.
Can anyone who is taking/has taken the program and who has a family and full time employment comment on how the time commitment fits into your life, both in terms of course workload and the overall length of the program? I keep finding myself looking at omscs, but I’m a full time swe and I have kids so the prospect of another major time commitment is a little daunting.
It can definitely be difficult working full-time with a family. I hit the halfway point in the program in Fall 2017 (after starting in Summer 2014) and I haven't been able to make any progress since then.
I know in order to do well in the classes I sign up for that I need to spend the needed time to complete the work, but it's been super difficult for me and balancing time with my wife and kids and my work.
I really wish they could launch more self-paced options for the courses, rather than keeping strictly to their on campus semester schedule. To stay active in the program I have to keep on signing up for classes, but then end up dropping them over the past few semesters.
On the one hand I want the masters degree, but on the other hand it's difficult to force myself to stay motivated and interested in the classes since not all of them are fun/enjoyable (or allow for the time to learn in a more self-paced manner).
I'm also not a big fan of online testing / proctoring...I would rather just complete projects ideally.
There've been other students that have been successful in the program even with full-time work, signficant others, and children in the mix, but those first few years in the program were definitely a sacrifice personally and it's been difficult to continue with that same level of sacrifice since the end of 2017 for me personally (hence my wish for a more self-paced option where I could accelerate/decelerate my own progress since the tie to the regular semesters really slows down my ability to make progress as someone that has difficult fully dedicating the time regularly now).
Check with your employer. You may be able to bargain some time off that you can dedicate to your education. I was able to get a full day off, and it's usually enough to do most of the work, if you're able to leverage some parts of your lunch/commuting to watch lessons. I have a job and two kids.
Remember that nobody forces you to graduate quickly. I always take one class AT MOST at the same time, so I don't get overwhelmed, and I alternate easier classes with harder ones, so I kind of "prepare ahead" for the harder class while attending the easier one.
OMSCS grad, father, and full-time SWE, here, also currently enrolled in the OMS Analytics program. I had my first kid mid-way through the program, and it was pretty rough. Something's gotta budge, and for me, it was sleep. In total, I pulled about 15 all-nighters, give or take, and watched a whole lot of lectures holding an infant. Most of the all-nighters were for large assignments/projects. It's tough, but can be done. Some tips:
1. Load-balance your courses. omscentral.com is an excellent resource for this, and I made sure that the workload never exceeded 20 hours per week (there's crowd-sourced statistics there). Pairing some of the more work-intensive courses with "easier" courses was very helpful, in retrospect.
2. Spread out your courses. It's self-paced! You can take 1 course per semester, skip a semester, etc. Before my kid arrived, I front-loaded and took a 3-course semester.
3. Watch lectures at 2x speed. This was clutch. Most of the content has very slow speakers, likely for clarity, so if you're a native English speaker, watching at 2x speed is usually reasonable.
4. Be okay with mediocre grades. This was probably the most liberating aspect for me. In undergrad, I really cared about my grades. Not caring about grades in OMSCS (or caring more about learning what I want to learn) was more liberating. There were many times I submitted sub-par assignments because they just didn't interest me, or I would rather have spent time with my family.
Hope this helps. If you attend the program, join the unofficial Slack group (I run it). Happy to chat with you offline.
You work your life around it. If you're only taking one course you can probably pull it off, otherwise you're going to have a bad time. There is a ton of lecture content, assigned reading, projects and tests -- much of it beyond what's available in the free udacity courses.
I've been taking two courses/semester and it's a significant commitment. 40+ hours per week.
First, this is not an ad. It's a post that was sent internally to students by the dean of the College of Computing in order to recollect the inception of the program.
Second, why should this be bullshit? I'm an OMSCS student, and others HNers are, too. Check the accounts: all of them seem in good shape.
I don't understand this attitude; if something gets praised, it must be a scam? Maybe it's just something good.
47 comments
[ 5.2 ms ] story [ 118 ms ] threadProbably an in-person MSc will be a bit better than OMSCS, but it will cost 7x-8x and would require people to move around the country.
There were 1 or 2 of my earlier classes that did feel phoned in, though (but did have strong TA involvement at least).
I haven't interviewed or applied for jobs since graduating, but it's worth noting that the degree itself is the same as the one awarded in-person. If you get involved in the community, the network is also surprisingly good; the slack channel has hundreds of students at all sorts of places.
I don’t know if my experience is considered typical. I worked as a TA in the algorithms course and then in Computational Photography while I was there. The algorithms professor was actually how I found the SV job I moved here for. I’ve seen it mentioned elsewhere already, but I can confirm that TAs run most classes—but it’s also worth pointing out that TAs in OMSCS are often comically overqualified, and 90% of the work in running a course is logistics (due dates, extensions, grading, etc.).
I think it’s as valuable as a resident program if you can’t justify leaving your current job or taking on debt to finance an advanced degree. Just make sure you try to learn everything you can while you’re in it.
Some of them are amazing, some of them are totally worthless.
Comparing it to my Master’s which has even less faculty involvement I doubt it. Watching videos and reading the textbook leads to the feeling you know something. Doing projects and preparing for exams, that leads to genuine learning. You learn far more working through past exam papers or doing all the questions in the textbook than you possibly could by trying to absorb information passively.
You would not get any of the class interaction, you would not get any of the projects, and you would not get any of the feedback from the projects. You would miss a ton of knowledge by missing those.
Also, the slack community makes the entire program -- you really learn a lot watching the people who know the material help people out through the projects.
A quick plug for my writeup: http://writing.maxrosett.com/reflections-on-the-georgia-tech...
It's the hardest thing I've ever done, and well worth it.
The quality of instruction varies wildly. OMSCentral is a student-run review site that covers the courses well.
The logistics depend almost entirely on the quality of the TAs in the courses -- there are courses like AI4R where the professor is heavily involved and answers all student questions, but the majority I've seen the professors are very hands-off and the TAs handle all student interaction.
https://medium.com/@mycahp/thoughts-on-the-omscs-program-at-...
- The instruction quality is comparable to free online courses (which is unsurprising since it's all delivered over Udacity anyway.)
- I learned a ton
- What Georgia Tech gets right is simple and a little sad: credentials matter. I learned a lot by going through a bunch of online courses and being graded, but a significant chunk of the value is just saying I have the degree. I wish that "I took a dozen MOOCs" carried the same weight as saying that you have an MS from a top-10 school, since in this case they really are exactly the same, but it doesn't. That's particularly true with arms-length interactions. Day-to-day my CEO cares more about the skills I picked up from the program, but when he's assembling slides about team for a pitch deck, the credential matters more.
Here's my full write-up: http://writing.maxrosett.com/reflections-on-the-georgia-tech...
Edit: Formatting
IMHO: the instruction (video classes) is comparable to GOOD QUALITY online classes. There're a ton of shitty online content.
And projects/exams are, IMHO, far harder.
But YMMV.
I'm finishing up my first semester, taking graduate introduction to operating systems (GIOS), and so far I'm super impressed with quality of the program (again, not sure how the other courses compare but I hear GIOS sets the bar quite high). In addition to the rigorous curriculum, the students actively chat with one another online; the current course has a very active Slack instance — paid out of pocket by one of the TA — where people chit chat pretty much every day, many of us sharing our frustrations with the projects that, as others have mentioned, drill the theoretical concepts into your head.
In short, I'm pleased with the program so far and look forward to the rest of the courses.
Overall OMSCS was a phenomenal value. The lectures are hit or miss, but that's not where the value lies in general, though there are exceptions (Joyner's classes like Human Computer Interaction). The value of the classes lies in the curated projects that high class professors designed and selected to facilitate learning. I learned so much in RL and ML especially due to those projects, as well as Educational Technology due to a project I selected myself.
FWIW I was working full time as a new software engineer when I started the program and after a year it helped me move internally into a data scientist position where I've now been for 2+ years.
My path was BA philosophy -> MS Finance (worked as an investment advisor) -> after a year break I went back for a post-bacc certificate in CS during which I interned as a software engineer -> OMSCS, which I started the same month I started my first full time software engineer position. A year later I transitioned to a junior DS position, and am now a DS. I'm 30.
Can anyone speak to the value or content of the Interactive Intelligence and Computational Perception/Robotics specializations?
It's run completely by the TAs, it does have one project at the end that allows the students to explore a bit outside of the basics.
That's the only one in that specialty I've taken so far.
I know in order to do well in the classes I sign up for that I need to spend the needed time to complete the work, but it's been super difficult for me and balancing time with my wife and kids and my work.
I really wish they could launch more self-paced options for the courses, rather than keeping strictly to their on campus semester schedule. To stay active in the program I have to keep on signing up for classes, but then end up dropping them over the past few semesters.
On the one hand I want the masters degree, but on the other hand it's difficult to force myself to stay motivated and interested in the classes since not all of them are fun/enjoyable (or allow for the time to learn in a more self-paced manner).
I'm also not a big fan of online testing / proctoring...I would rather just complete projects ideally.
There've been other students that have been successful in the program even with full-time work, signficant others, and children in the mix, but those first few years in the program were definitely a sacrifice personally and it's been difficult to continue with that same level of sacrifice since the end of 2017 for me personally (hence my wish for a more self-paced option where I could accelerate/decelerate my own progress since the tie to the regular semesters really slows down my ability to make progress as someone that has difficult fully dedicating the time regularly now).
Remember that nobody forces you to graduate quickly. I always take one class AT MOST at the same time, so I don't get overwhelmed, and I alternate easier classes with harder ones, so I kind of "prepare ahead" for the harder class while attending the easier one.
1. Load-balance your courses. omscentral.com is an excellent resource for this, and I made sure that the workload never exceeded 20 hours per week (there's crowd-sourced statistics there). Pairing some of the more work-intensive courses with "easier" courses was very helpful, in retrospect.
2. Spread out your courses. It's self-paced! You can take 1 course per semester, skip a semester, etc. Before my kid arrived, I front-loaded and took a 3-course semester.
3. Watch lectures at 2x speed. This was clutch. Most of the content has very slow speakers, likely for clarity, so if you're a native English speaker, watching at 2x speed is usually reasonable.
4. Be okay with mediocre grades. This was probably the most liberating aspect for me. In undergrad, I really cared about my grades. Not caring about grades in OMSCS (or caring more about learning what I want to learn) was more liberating. There were many times I submitted sub-par assignments because they just didn't interest me, or I would rather have spent time with my family.
Hope this helps. If you attend the program, join the unofficial Slack group (I run it). Happy to chat with you offline.
Bs work, I don't care about GPA, I'm trying to learn and get the degree at the end.
You work your life around it. If you're only taking one course you can probably pull it off, otherwise you're going to have a bad time. There is a ton of lecture content, assigned reading, projects and tests -- much of it beyond what's available in the free udacity courses.
I've been taking two courses/semester and it's a significant commitment. 40+ hours per week.
Second, why should this be bullshit? I'm an OMSCS student, and others HNers are, too. Check the accounts: all of them seem in good shape.
I don't understand this attitude; if something gets praised, it must be a scam? Maybe it's just something good.
I understand the cynicism but sometimes there are just good things in this world. For all of its warts, OMSCS is one of these good things.