So maybe things changed, but Stanford's MSCS program used to be a full time intense and super legit program with a high bar and wasn't a cash-grab. Many schools do have a lot of international students paying a huge amount of money so they can get a degree and F1-OPT visa to work in the USA.
I actually would not say the same about some of the other big name schools who do have programs that are less renown as their undergrads.
EDIT: Yeah I think this is still true:
"Q: Does Stanford University offer a part-time CS master's program? A: Stanford University offers a part-time masters program called the Honors Coop Program (HCP). The program is available to students who are pursuing an MS and are employed full-time. HCP students can be either remote or local. Please check out the web site for the HCP program at [http://scpd.stanford.edu/programs/masters-degrees](http://scpd.stanford.edu/pr…
Can you elaborate more about your goals post graduating? There are two buckets of programs you can look at: bootcamps (which are meant for people with almost no past experience or training and helping them get a first job) and career accelerators (focused on practice and interview prep and levelling up your career).
If you have a CS degree, both buckets could work depending on those goals.
If you feel like you learned very little during your degree, have no internships, haven't had any interviews ever, a more advanced bootcamp might be good.
If you feel prepared for a job but have no prospects, are getting but failing interviews, a career accelerator might be good.
In terms of deferred payment, some programs have ISAs and some have loans that you can pay over a long period of time.
Pair programming is key. The interview is basically live coding with someone and you have no idea how different things feel when coding with other people hovering around. On the job for real, you don't often do live coding like this, but doing it forces you to confront topics you THINK you understand but you don't actually understand in a way that you can explain to other people haha.,
Why do you think people fail to get in?
Also for that question, if they expect you to pass in an index to an array recursively that doesn't show a strong understanding of recursion and how it would be used to solve complex problems in the future, like DP. So I would barely call that testing recursion. That is just "can you write a for loop recursively".
I hope they properly teach recursion in Codesmith afterwards and hwo to use it to solve complex problems. At Formation, recursion alone is something most people spend two to four or more weeks of intense practice to truly understanding and weild as a powerful tool to solve problems with.
Most people at Apple are huge fans of Apple products. I would probably go there and get settled in and then revisit in a few years.
Your passion for Apple products will help drive your work.
Imagine working at Roblox where some coworkers have that same passion for Roblox and you are like meh it's cool.
AR/VR is a great space. I think it will set you up long term.
Absolutely, I'm specifically talking about Silicon Valley based top software companies which happens to be considered the top places for engineering culture, and tech driven companies.
NYC for example skews a bit more ivy league and is more Finance driven companies.
Both are great opportunities, I know many people at both and both are incredibly different.
Good news is that both are great on a resume for the future. Bad news is that it's a big tough decision to make.
Apple: it's pretty hard to get into Apple because each team hires differently and is fairly independent. People tend to stay a long time on their teams. You can move, but people tend to get really good at what their team does and have less breadth.
Roblox: hot consumer product that changes rapidly. The team has absorbed a lot of people from other top software companies and is someone more representative of Silicon Valley.
If you have a lot of initiative and drive, and/or love Roblox as a product, I would choose it and get ready for ups and downs. If you want a stable and great steady career without any curve balls, go Apple.
I mean UIUC is a great school. The difference is the top companies literally target those top four schools with extra resources and their best recruiters, so you have people fighting over you. The targets/goals for recruiters for hiring from them are also higher.
You'll also find that almost all of your peers will go to top hot tech companies that will help you later on in your career.
Again, nothing wrong with UIUC but I see a lot more Stanford on resumes at the top companies in Silicon Valley.
LinkedIn if the person has never lived in Silicon Valley and gets to go in person. Will be a big eye opener.
My brother worked at Riot for 10 years and I would go there if you are obsessed with any of their games. It will make for a fun experience and lifelong friends who feel similarly.
Both are good but if you want in on leading software companies in Silicon Valley, 100% Stanford. You'll easily interview at all the top companies and get offers on the high end of the bands. Seen it first hand how much effort recruiters put into Stanford, MIT, Berkeley, and CMU.
I actually think while both options are great, I might lean Snap despite the troubles. You learn a lot in hard times, sometimes more than you did in the good ones.
There are a group of people saying that recursion and OOP are.now always tested to get in. We work with Codesmith alumni who still need recursion practice after graduating so that seems like a high bar to me.
If anyone has info on the consistency of the bar, please comment. Does everyone get the same questions/topics?
I can give you the Formation analogy now haha. We put you in the shallow end without floaties (we don't lecture) and stand behind you and continuously push you towards the deep end. How hard we push you keeps changing depending on how much you push back each step of the way and then eventually you are in the deep end all alone. Each person has a completely unique pushing needs/pattern.
I don't know. So the underlying math and practice data sets can be for the right people, but it's hard to simulate the sheer amounts of data people have at top companies. But the way Codemsith students exaggerate a 6 week OSP project into legit work experience, if they are banking on doing the same thing and turning a 6 week project using a public data set into a work like experience well see how it works. I think it's easier to fake it with full stack engineering where there are ten of thousands of stacks people work on. You can't fake PHD level math.
I think there is a lot of value in this model for people that are already strong to make them better, so it's not a waste for these people but it might limit the ultimate breadth of skill levels of people Codesmith can support long term, which might be totally fine. They are expanding to machine learning so if they can just cherry pick the best people in more and more disciplines that can scale too... it just changes the goal from "I want to get into Codemith so they can develop my skills to be successful" to "I want to get into Codemith because it validates I have the traits needed to be successful based on similar people". Both can work.
Surprisingly for all I know about Codesmith, I don't know much about the actual lecture contents... but this sounds really bad. It sounds like it relies on the raw abilities of the people both technically and communication wise with peers to learn.
If someone was really far behind, even with 1-1 peer time via a fellow, how could they possibly catch up with this model?
No, and I don't know much about them unfortunately. They were acquired by Stride, a big education company at the end of 2020 but seem to haven't changed much since then.
Sure, so people are reporting the entrance bar getting more selective. For example, basic recursion is required, and OOP is required.
So there are a fixed number of slots, and presumably more applicants than can fit, so they are choosing stronger and stronger applicants.
They have added an in person cohort and have 4 cohorts now on a 7 week cycle. So they have been adding capacity, but presumably not as fast to server everyone.
So what this means is the incoming people will be more experienced and stronger than in the past, and will get better outcomes - both speed to outcome and compensation.
That's an interesting perspective, thanks for sharing. I didn't realize that would push people away from the industry.
Codesmith has a bit of a cult-like following and branding (that is also pushed internally constantly - always told the curriculum is perfectly optimized, the resume process is perfectly fine tuned, etc...) and I guess that comes across on the outside too - 'this CSX course is perfectly prepared for beginners and do it to get into codesmith and since codesmith is the best, then....'
It's definitely not that the best. Their content team is working on a story about people who do CSX and get "junior roles" immediately without doing Codesmith, as opposed to the "mid level and senior" roles you get at Codesmith. So I think they internally think highly of CSX if they think CSX is equivalent of 'other bootcamps'. Watch out for that blog post haha.
Do Apple and Amazon allow you to run a company or non-profit company in the technology space while also working there full time? I know Facebook did not allowed this while I was there unless you cleared it by the competition board. Or did the founders leave those jobs already?
Geography is a big factor. Tech Elevator is an a lot of lower paying markets, like Cincinnati, Cleveland, Wilmington, Detroit. Those markets pay 20 to 30% less, even at the same companies. e.g. Amazon Detroit entry level is like $115K versus $160K in NYC.
So $120K -> $84K.
I agree that I should be more precise and appreciate the pushback and I did a deeper dive. A larger problem is most big tech do not allow you to have another W2 job in the tech field because their contracts cover all your IP created in the tech space and the OTHER W2 job usually has a similar contract and hence the conflict.
Agree on 52 Fellows, 3 months W-2 @ $1K per week.
17 prep instructors who have full time jobs elsewhere (I don't know the compensation, it's not gift cards but I don't what it is and I would somewhat expect it to be something hourly yeah, I shouldn't bucket them into the 50K a year - I just don't know but I didn't think it was equivalent to the full time instructors and have ZERO basis to believe it's lower, just guess)
15 engineering mentors who have full time jobs elsewhere (these are the people who I thought were not paid W-2 but paid with gift cards - or equiv…
I'm including alumni who do part time instruction in there too, and alumni who do mock interviews for free or gift cards (that number might be much higher actually)
Going off the about page. But there are also several people on LinkedIn who claim to work at Codesmith not on the about page so no idea how large this number is.
I think it's very unlikely they will sell. They have a good thing going and are printing money right now because it's 100% upfront. They have about 80 former students on staff who are paid like 50K a year or with gift cards and have scaled that very well.
I think the only risk is if they scale too large or too broad.
They benefit from being very selective and their weaker students don't have as much success. So they will either have a cap on how many people meet their bar, or they will have to invest in training people beforehand to get them to the bar.
I also think it's a risk that someone catches on to the exaggerated work experience issue. I haven't talked to one non Codesmith industry person who thinks that's ok and if they get large enough and people catch on it may not work anymore. I know this is controversial.
They are experimenting with a machine learning cohort and we'll se…
Yeah definitely would love to know who has experience going in versus no experience. They continue to raise the entrance bar to people who know recursion, OOP, etc... I'm also curious because I thought they are a bootcamp for people starting at a low level.
CIRR's "percentage employed in 180 days" is accepted offers / graduated, correct. And is 100% misleading not accounting for people who don't graduate.
As I said above, "Percentage of job obtainers who reported salaries" also lets you remove people from the salary counts, which still including them as placed. Why would this happen? I'm not accusing Codesmith of this, but Tech Elevator is 100% and I don't know why Codesmith is missing salaries for people counted as placements. My theory is people who ghost but can be confirmed as employed from trustworthy data.
Ah yes, the CIRR ambiguity and you have to do some math on your own.
So the only hard number is "Number of graduates included in the report: 91"
The rest of the numbers are all percentages and honestly the numerators and denominators aren't specified super clearly in the reports. Even the CIRR spec isn't 100% clear.
"Graduates included in report" in their worksheet is based on a column "Included in placement data?" which is an AND formula of 1. intended to get new job within 3 days of starting program and 2. has work authorization 3. "effective" graduation date was within the six months window of the report. So I'm interpreting that this actually means someone graduating late, might be included in a later report.
So my formulas above are estimating when someone signs a contract what percentage graduated on time and then getting a job within the expected 180 days. Basically trying to…
It means that a lower percentage of people were placed within 180 days compared to last half. It's more reflective of the job market I think than Codesmith because it's not that much different.
What it does mean is the salaries are of the people getting placed only. So higher salaries across a smaller number of people.
NY H12021: 94.7% graduation -> 89.4% placed in 180 days = 84.6% people starting getting placed
NY H22021: 92.8% graduation -> 83.6% placed in 180 days = 77.5% people starting getting placed
LA H12021: 94.8% graduation -> 85.2% placed in 180 days = 80.8% people starting getting placed
LA H22021: 91.9% graduation -> 80.9% placed in 180 days = 74.3% people starting getting placed
More CIRR H2 2021 results out! Codesmith included - with a lot to unpack! Overall lower placement rates and much higher salaries of those placed, and a few more fun things!
Where to begin, so much interesting stuff in the part 2 CIRR results, focusing on Codesmith as it's the most interesting to this subreddit.
Overall interesting notes (incomplete but tried to select a few things I found interesting):
* By far the most interesting thing in East Coast (EC, formerly NY), salary growths were super large. Median is up to a WHOPPING $140K but more interestingly in H1: 18% of people had salary $140K+ and now \~52% of East Coast grads made $140K+, that's a massive shift! And similar to FTRI there is a dip to almost zero in $150 to $160K with \~20% in the $140K to $150K bucket and \~20% in the $160K+ bucket. So my hypothesis on this adding in my industry knowledge is that \~10 to 20 people g…
Will Sentence, the founder, is more opinionated about pedagogy I believe (never talked to him directly) and is a really smart person who puts a lot of effort into teaching effectively.
But I agree that CSX is pretty weak. It's intended as a "top of funnel" to Codesmith and many bootcamps follow this approach. They want you to join CSX and work with some peers on Slack. Then like it enough to sign up for a few hundred dollar prep course, taught by alumni. Then like that enough to join the immersive.
Most bootcamps have a similar approach to this. CSX isn't really intended to teach people anything as a standalone tool. Most of the people I know who got into Codesmith used many tools including Leetcode "easy problems" to get in. They want you to attend free seminars and pair programming sessions where they can build a relationship and help the right people make their way to immersive and…
Codesmith is somewhat split on outcomes. There's one contingent of engineers who have experience, don't really need the curriculum and benefit from mentoring others, the accountability, and the network. This group is aiming for top tier jobs. I believe this represents about the under 40% of people who make salaries over $130K.
Then there is another contingent who end up making under $110K which is about 20% plus I would include the people who don't get jobs in 6 months (another 10%) and then the people who don't graduate (5%) (all this based on 2021 CIRR reports). I'm making a BIG ASSUMPTION that the people who don't graduate and don't get jobs are on the less experienced side but I'm sure there are people for whom this isn't the case.
Anyways, that means APPROXIMATELY BASED ON CIRR 35% of people who are weaker or have no experience get jobs under $110K or no jobs within 6 months of g…
These are our last \~30 offers accepted in order (excluded declined offers obviously): Square, Flatiron Health, Amazon, Amazon, Lockheed Martin, WePay, Amazon, Visa, Doordash, Virgin Orbit, <top startup redacted>, UIPath, American Express, <mid size not top tier redacted> Amazon, Square, <top startup redacted>, Amazon, American Express, Zapier, Edelman, Klaviyo, <top startup redacted>, Dialexa, Amazon, Google, Plaid, Sense, <not top tier startup redacted>, <top startup redacted>, Figma, Google.
I think this list demonstrates both consistent top tier placements, as well as the breadth of placements on an individual basis. We still don't have more salary data to give because we don't want to give anything out that we feel doesn't communicate what we're doing.
As stated before, the methodology for the data is in the fine print of that dialog on the website and it's very clear how we compu…
Yeah I totally think the percentages are very clear and the 20% under $110K and 20% over $140K is a lot more interesting than the median in the middle but I meet a lot of people who almost always say "but Codesmith grads make $125K" and the data, which is very different from all other programs that report, looks more to me like people with experience have made $140K+ (which is actually market rate a mid-tier companies for people with experience) and people with no experience tend to make under $110K. Most people in this Reddit have no experience, but they believe they will make $125K. I wish they had data for experienced and non-experienced people but that's not part of CIRR.
You are straight up wrong about CIRR. The median salaries are of people who reported salaries and got jobs only and exclude everyone else. So drop off 1 is the graduation percentage. People who don't graduate are not included in the median salary calculation. Then there's people that don't get jobs. They are not included in the median salary calculation. And then outside sources can be used to confirm employment for percentage employed but without a salary to include the median calculation. You said you do a lot of research so read the CIRR and it's very clear in the Excel worksheets they provide... complete with examples showing exactly this
This is exactly what I'm talking about how CIRR has built in vagueness to make this happen and help paint programs in the best light.
I tried to give some reasons above but I don't even like what we have now on the site, because the outcomes vary so so so much by goal and we have to very carefully decide which numbers best represent what we do - which we haven't done yet - so we encourage people to talk to whatever random Formation person they can find with similar backgrounds and chat with them. Like we literally have had 10+ offers over $300K TC in October for whom that is the market rate and we help the people choose between offers. The uniqueness of those situations for those people are not captured in an average salary metric because and we want people to know they are getting a bespoke experience aimed at helping the accomplish their goals.
I've said this before but some of Codesmith's alumni get stock and bonuses as well, excluded from CIRR, and CIRR does not paint an accurate picture of their outcomes as well.…
You’re asking to compare people that get jobs within 90 days of starting so you can compare it with people who get jobs within 90 days of FINISHING. Makes no sense. You are asking to compare people that do 50 hours a week to 10 hours a week. Makes no sense. You are asking to compare people with a goal of getting a job in 8 months with those who want one ASAP. Not everyone wants a job ASAP at Formation. People often get several offers and keep training until they get one they like. All of this makes adds to why it makes no sense to give those CIRR numbers.
I agree with your ask though. I would love to provide better data that can help people estimate how long it might take to get a job given their unique circumstances across over 5 dimensions. So far it hasn’t been a big ask and people talk to current Fellows to get an idea of how it works. Quite frankly we are doing ourselves a disserv…
I keep saying the same things but everyone comes in at different skills. We have someone with literally zero experience who got a job af Amazon entry level and a staff engineer with 6 years of experience. Aggregate times don’t make sense. They make sense if everyone is starting at the same point like a bootcamp or school. They take you from A to the best B possible. We take you based on your B goal and if we think we can get you there from wherever your A is. It’s seriously nothing like a bootcamp or anything remotely similar.
Im here to give people advice because I work with so many past bootcamp grads and see a lot of misunderstandings. I’m sorry if my involvement here is confusing but I would ask you try not to read into it.
Again, not a school, no graduates, no end of program, no curriculum and we can’t be compared to a bootcamp head to head. All good questions to ask and also exactly why we can’t possibly report CIRR like metrics.
I think you have a big misunderstanding about what Formation is and it might be helpful to research more instead of trying to make us for your mold for what you think we are.
I feel personally attacked because I repeatedly tell you we don't have a school and we are running something different, and you repeatedly call Formation a school or a bootcamp and repeatedly ask for data the schools and bootcamps offer. You seem so fixed on the fact that you believe we are a school and I've offered at least once to do a call with you to explain what we are. Yet you keep saying it and the I have to keep correcting it.
Oh so all your projects are under one section? Like a single top level item “Open Source” with your OSP and your all projects all as equal sub bullets? That’s my recommendation as well yeah. Very curious to see why the majority of resumes end up with OSP as a separate item! Let me know if you get any insight why.
Always appreciate good discussion Ben and Triathlete!
How many people don't have jobs yet though. One key problem with CIRR is that it is focused on the outcomes of people who get jobs and report. It's not the median salary of everyone who started, only the median salary of those that got intended to job hunt, graduate, for a job, and report a salary.
So this might sound very subtle. So assume that no one will read the bullet points and the key trick is having the following
OSP:
-bullets
Open Source Projects:
1. solo
2. refine
3. enhance
By putting your other projects under a single "open source" bucket it tricks someone into thinking the OSP is likely work experience (even if it says open source somewhere.... people don't read it)
Which companies hire who and when will change over time. We had a lot of people going to FB and Google and then they froze. Now we have a lot of people going to Square/Block and Amazon. I suggest that no one follow the current week to week trends and plan based on that.
Yeah actually I do. The world is very large. These companies are very large. People see hundred and hundreds of resumes.
1. Several times a week people on my team mislabel Codesmith alumni as industry experienced based on their LinkedIns. Recruiters spend seconds looking at your resume and they don't read bullet point 15 that says "product incubated under oslabs" and when they do, they aren't pondering what that is and if it's open source
2. I asked some people the other day about this with some examples, industry experienced people and two said expletive laced sentences about this practice.
You can blame it on the recruiters or the companies but things are the way they are because the vast majority of people have integrity in their resumes and don't do this and companies don't build teams of recruiters who are trained and focused only on this tiny edge case.
The problem here is tha…
So there is such a size-able Codesmith contingent at Capital One, they have their own Slack channel and they can refer people to a variety of teams.
Capital One has a variety of positions, but the one most people are getting is "Software Eng - Senior Associate" which pays around $150K a year base salary and total comp. A FAANG entry level is about $200K+ total comp based on performance for comparison.
Reasons how this works.
1. They only have one level lower than this that is very entry level "Associate Software Eng" and it's meant for new grads and kind of like a mini internship. So anyone with any experience would be considered for "senior associate"+.
2. Some of these people at Codesmith have experience already and don't do anything special to be considered.
3. Some of these people at Codesmith list their group projects as "work experience" and mislead the company into thinking t…
Yeah it has solid cash compensation too, I think it's a great first job for someone with no experience.
Oh one followup: "Senior Associate" there is like "Early Career L3" at Google. and "Master Engineer" or "Lead Engineer" is what Google calls Senior Engineer... just showing the level names being meaningless as it's a good example of that.
From my knowledge working with a wide variety of Codesmith alumni I disagree that "they don't teach anything". I see two buckets of people: first are the 2/3 of people with zero experience who self taught enough to get in, and they learn a tremendous amount of practical skills; second are the 1/3 of people with experience who do say things along those lines, and it's likely true because Codesmith's is a bootcamp aimed at helping people with no to little experience. The 1/3 of experienced people probably shouldn't go to Codesmith to learn skills but the 2/3 of not experienced people do find it incredibly valuable.
At Formation (disclosure: co-founder, not a bootcamp, work with experienced engineers) we have seen a slightly increased demand from bootcamp grads who can't find jobs, and our outcomes remain very strong, but we are targeting top tier companies and it's not for everyone. There…
Just my personal opinion, but this post makes it sound even more like a scam than the original post even though I can tell you don't want it to be :(
When you say above that you attempted 3 other cohorts and they all failed and you refunded everyone, I think that's great honesty and transparency but it doesn't give a lot of confidence that your new instructor will do better - you are risking a lot on one employee here showing up and doing a good job.
Finally, one instructor for NINETY PEOPLE is an insane ratio, I don't know anyone that can have that kind of ratio and deliver quality. And you are saying you will have at least 90 people with one instructor.
I honestly appreciate your effort and drive, but maybe start off smaller with like 10 people and get it working and then expand?