top | item 25415017

Lambda School is the biggest mistake I made this year

602 points| barry-cotter | 5 years ago |reddit.com | reply

431 comments

order
[+] soneca|5 years ago|reply
What is sad to me looking from the outside it is that it didn’t have to be that way. Most of negative reviews seem to boil down to too many students, not enough support, not professional enough course management. Symptoms of growing pains. But why Lambda School had to grow that much that fast? It didn’t. It shouldn’t. It was at the expense of their students.

Lambda School seemed a great idea from the start, revolutionary even. And the investors (pg mostly) would praise how insanely ambitious the founder was. Then I started following him on Twitter and saw that. He had that drive to be the Facebook of education, the slack of boot camps. But education doesn’t work that way. A new Facebook user can start using it 4 hours a day from the first day and get all the value they expect. A student can’t. Students need time and support to learn. But Lambda and its founder couldn’t wait for that if they wanted to become a unicorn.

[+] austenallred|5 years ago|reply
CEO of Lambda School here.

This is disappointing, of course. We have 180+ people working really, really hard to make Lambda School the best experience it can be. At times we’ll fall short, and, as another poster pointed out, the margin for error is just razor thin; thousands of happy and successful students can be out there, but it’s the ones who don’t make it or have a good experience that cause us to lose sleep (and make it to the top of HN).

We take our responsibility in that regard very seriously; students come to us to change their lives and do something really, really difficult, and anything we do affects people in a very profound way.

To the parent’s point, in this particular instance I don’t think this student's frustrations are the result of too much growth or lack of resources per student. We haven’t really grown over the past year or so, and the school is still very well resourced. We have just north of 2,000 students and ~180 full-time staff members, the majority of which are student-facing. Those may be suboptimally distributed (they are, and always will be to some degree) but a cheap education as a result of aggressive cost cutting this is not. As such, the solution is much less exciting than “need to spend more money” or “need to grow more slowly.”

This student’s frustration comes from a series of changes we made simultaneously (something we avoid doing when we can). Some of these changes worked very well. Some did not. And this is the tricky part: some of the changes that a subset of students did not like are unquestionably for their benefit. So we’re working through those changes as well as debates around the trade-offs around, “Do we want our students to love us or be successful?”

The specifics of what students didn’t like: The first is a mentor/mentee program we set up within the school. Employers tell us literally every day they want students who truly understand the development and release cycle, and work on teams. We created a mentor/mentee program where every student has a chance to do that, and the result of that program so clearly creates better outcomes that employers ask specifically for students who have done that.

Some students love it. Some students see it as one more thing distracting them from writing more code, and believe the only thing that will determine their career success is ability to write code (as you all know, ability to write code is necessary but not sufficient to success as an engineer). We’re digging into the details of how and why it fails when it does. I don’t yet know what we’ll do there but the solution is likely more quiet and subtle iteration to the student experience.

The other change was our slowly winding down of the Team Lead program. We used to have a program where we would hire some students on short contracts to serve as something akin to TAs. Sometimes this worked very well, often times it didn’t work well, and the experience was brittle and unreliable, so we shifted that budget back into full-time roles. Doing so has revealed gaps that TLs were plugging that should have never existed (the majority of which should have been solved by software and not people from the get-go.)

So, Lambda School is not yet perfect and likely never will be. We’re properly resourced and not growing too quickly. We’re trying out best to make everyone we admit happy and successful. Turns out that’s extraordinarily difficult to do, and we’re doing the best we can, and I’m always open to any feedback on what we can or should do better.

[+] javaIsGreat|5 years ago|reply
I dont know anything about Lambda's failures to serve students other than most students are upset at paying for less than steller experiences (most of the other prestigious bootcamps share this), but my gut tells me I dont think they grew to fast, its just the calibur of student went down, i feel like the hot new bootcamp of the day is able to get really bright and/or qualified students. More average learners require more time and resources and the nonspecial sauce will NOT get them a dream job in less than a year which is the opposite of what all of the marketing leads u to beleive for all the bootcamps.

Lambda seems to have been trying to branch out to more normal learners.

Students that dont share those bright/already qualified credentials, do benifit a little bit of a "rub" effect by being associated with those bright/qualified students, and get to study/work with them during the course of the bootcamp so everyone wins in the short term. For average learners without significant backround IT knowledge or a prestigious degree, realistically they are going to need more than 3-6 months or a year to get a cushy programming job in the US (which seems to be what all of the bootcamps advertise they can offer students).

I consider myself an average/below average speed learner, took a bootcamp, i was surprised to learn the guy i was pair programming with had a CS degree from MIT lol. Some devs already had work programming experience, but most of us normies benifited by pairing with them and talking to them about problems/code/computers/study habits.

[+] jgalt212|5 years ago|reply
> praise how insanely ambitious the founder was

It seems to me that VC's often link ambition with intelligence and other assorted skills needed to get the job done.

[+] yanslookup|5 years ago|reply
My original comment was too mean spirited, even for this alt account. I think what I meant to say is, Austen Allred does a good job of talking his book on Twitter. He also seems like a great Dad and partner.
[+] michaelcampbell|5 years ago|reply
> praise how insanely ambitious the founder was

If this isn't a warning sign of a spectacular failure, I don't know what is.

Not to mention praising it, after all the ink spilled on "rockstar" "ninja" types and how bad they are, yet they still get kudos and praise.

If you ever have a manager that espouses the "we don't want the heroic cowboy type here", run, because that's EXACTLY what they want.

[+] Siira|5 years ago|reply
Perhaps their success rate has gone down, but their absolute successes have grown enough for them to accept it. Perhaps they should add some more expensive tiers for better student support.
[+] airdweia|5 years ago|reply
Please please please under no circumstances attend Lambda, everything the poster says is true. Two weeks before the end of CS for me they initiated a pass/fail for a test that we were never taught anything about and then mandated that you repeat the entire CS curriculum if you weren’t able to pass it. They said that it didn’t matter that I passed all of my sprints and scored 100% on the unit assessment, that I needed to repeat a class (along with 3/4 of my cohort) just to pass this test. They never owned up to the fact that they didn’t teach the material on the test, as was evident by the fact that they removed ALL GitHub projects in CS and replaced them with Codesignal challenges. I messaged “Student Success” for weeks and was ignored and when they did deign to respond to me it was a generic canned response that addressed none of my issues save “Deal with it”. I took to speaking about the problems in the open channels and I was removed from Slack and forcefully “withdrawn” from Lambda. It is utter chaos there.
[+] dikeledi|5 years ago|reply
I got drawn into Lambda School at the peak of their hype early last year and considered myself extremely fortunate when I got admitted into one of their pilot EU classes. Suffice it to say that the reality was...significantly poorer than advertised. Not only because of the much-discussed reasons which are true, but most importantly for me because they lied about having hiring partners in Europe when they didn't, and left the students to fend for themselves after graduating. Most of the students saw this happen with earlier cohorts and there was a steady trickle of dropouts while I was there. To top it off, around October last year, LS suddenly announced that they would be "pausing" the EU programs indefinitely to "focus on improving the student experience" which seriously drained student morale. I'm pretty sure no more than 35% of students who enrolled to Lambda EU have jobs currently and since they're not as loud as the Americans, their plight isn't really highlighted. Most of the students who managed to get hired either did it on their own, or were previously experienced and came to Lambda to improve their prospects. I managed to get a job through my own means, but would unequivocally consider enrolling to LS a mistake, and it's telling that they've not bothered to follow up with me for ISA payments.

Addendum: This was the kind of delusional fever dream perpetuated by Austen that kind of attracted me to LS last year:

https://www.coursereport.com/blog/lambda-schools-europe-inco...

https://timesofindia.indiatimes.com/people/after-us-india-wi...

[+] jacobion|5 years ago|reply
The thing about efficient markets is that lucrative activities which don't produce much value—borderline scams—can block activity which would be highly wealth creating. For example, there could be a huge untapped retail market for financial derivatives such as interest rate and fx hedges. However, it's very hard to access this market because historically anyone who has started selling such products has pivoted towards targeting degenerate gamblers in order to fleece them. Normal people who want to hedge conclude that it's better to live with the risk than to be the sucker.

Here we can see that there are lots of people who are prepared to make significant personal sacrifices to work in the software industry. The industry itself is crying out for more people and new people. But it literally doesn't make economic sense, either for universities or for new players such as bootcamps, to train people properly. It's vastly more lucrative to confuse them and then rip them off. It's hard to see a simple market mechanism which can resolve this, as opposed to rebuilding something similar to the gatekeeping and signalling functions around universities, commonly regarded as broken nowadays.

[+] rswail|5 years ago|reply
Learning to program is more like an old fashioned trade than a profession. I don't mean that in a derogatory way. I mean that what should be happening is apprenticeships.

If "industry is crying out for more people", then it should be investing in that in the same way it does in every other part of the supply chain.

They should be hiring and training programmers, not externalizing their costs.

[+] Benmcdonald__|5 years ago|reply
Same as health care. Without regulations the industry would be full of snake oil. And yet the regulation kill innovation in the industry. Money all devolutes to scam. Look at digital money. Not even HN community can spot the "crypto" scams
[+] specialist|5 years ago|reply
> scams can block activity which would be highly wealth creating

What's this phenomenon called?

"Spoiling the market"? "Shitting the bed"?

Basically, when inauthentic activity so increases transaction costs that the authentic market is harmed.

My aunt once had a pretty good cottage business selling bespoke wall paper stuff on eBay. Over time, inauthentic entries crowded out her real product. SEO, fakes, knockoffs, misnaming stuff. Her customers could no longer find her products thru eBay, even when using exact match search terms.

Sound familiar?

Using terms like "fraud" and "scam" seems grossly inadequate. I'm provisionally using "authentic" vs "inauthentic", which suck, for lack of better terms.

A parting thought: Somehow Amazon managed to become worse than eBay, for both vendors and customers. Amazon boosts their own brands. And has a separate payola lane for "sponsored" entries.

[+] 3np|5 years ago|reply
> For example, there could be a huge untapped retail market for financial derivatives such as interest rate and fx hedges. However, it's very hard to access this market because historically anyone who has started selling such products has pivoted towards targeting degenerate gamblers in order to fleece them.

Tangential, but this is pretty much emerging in the cryptocurrency/digital assets/decentralized finance (defi) space. To your point, out-right or borderline scams are a-dime-a-dozen there right now and there are plenty of incentives to cut corners in order to get growth, but there are a lot of ethical and promising projects as well.

[+] ghufran_syed|5 years ago|reply
How exactly are students being ripped off? They pay nothing up front, and only pay lambda school afterwards if they are financially successful in a tech job.
[+] Grustaf|5 years ago|reply
Just because it’s more lucrative to rip people off doesn’t mean it doesn’t make financial sense to run a good code bootcamp. That’s like saying it doesn’t make sense to work because it’s more lucrative to steal or live off welfare.

But sure, it’s a common problem in a market economy. Presumably issues of information asymmetry should be alleviated in the days of the internet.

[+] Mikho|5 years ago|reply
Happens when a business model is to sign up as many people as possible for the ISA (Income Sharing Agreement) and then immediately monetize that ISAs by selling them with a discount to a 3rd party.

That's basically Lambda business model. The main part is to sign up as many people as possible since in mean hard cash right away for every account that became locked in the debt.

Education is just an irrelevant theme-trick to make people owe money. That's classic sting of when a mob seduces a target person to play in a rigged gamble game just to make the target owe this mob some money.

[+] TheRealDunkirk|5 years ago|reply
> ... then immediately monetize that ISAs by selling them with a discount to a 3rd party

Is that really true? If so, this is shocking to me. If LS sells the ISA, then, obviously, they are no longer "aligned" with their students' success. What's the point of their "revolutionary" arrangement then? This makes perfect sense out of the jumble of anecdotes I'm reading.

From further below: https://www.theverge.com/2020/2/12/21135134/lambda-school-st.... In other words, yes, this is what's happening, and it's a disaster, for exactly the stated reason.

[+] bertjk|5 years ago|reply
Wait.. so they were securitizing their ISAs? I had no idea they were doing that. It is a brilliant twist I didn't expect, and that also causes me to lose all respect for them. I liked the ISA idea before, but this... is perverse.
[+] texasbigdata|5 years ago|reply
Anytime you securitize an asset like that credit quality gets heavily scrutinized. From a working capital perspective you can cause a real liquidity crunch if you’re simultaneously a) growing and b) deteriorating factored asset quality leading to lower advance rates on subsequent tranches in the SPV. Restated in English, if you’re taking payday loans and your income is declining you’re about to have a bad time.
[+] thinkingemote|5 years ago|reply
"I signed up for 18 months of programming... They suddenly reduced it to 12 months. As of yesterday, I've withdrawn after 4 months as this is not the program that was sold to me. I completed 22% of the program...but under their new reduction, they said I owe 37%"

Seems like they would have a good case for getting their money back, but might need to consult a lawyer.

[+] gnicholas|5 years ago|reply
Well, it's even better for the student because the tuition is paid via an income-sharing agreement. That means you don't have to try to "get your money back" — you just have to defend yourself if they actually sue you for a share of your earnings.

This means that Lambda would have to deal with the bad optics of "suing their students" (which sounds worse than simply refusing to give refunds).

If Lambda did actually sue, students would have to defend themselves. But it's always better to be the party that starts out having the thing at issue (i.e., the money).

[+] blntechie|5 years ago|reply
Most open world reviews i.e. non curated ones I have seen for Lambda school have been negative to at best neutral. I believe they have a place in the education market if they do it right. Model is good enough but execution appear to be severely lacking.

It’s interesting that pg is a big fan though.

[+] jstummbillig|5 years ago|reply
The biggest mistake ls could make is not recognizing how little room there is for (perceived) failure with their product. It is among the most impactful, personal products you can buy online (considering life time and money investment).

Customers will understandably be in a vulnerable place. You are not exactly selling someone a boat. You are quite literally selling tech savvy people their future and they know how to use social media. Those who are brave enough to voice their troubles with education in the open – which is a stigma, maybe so than ever – will have no trouble finding an empathic, educated audience.

This is akin to treading in self-driving space: Every player is super aware of the non-existent room for error. The whole space has constantly been 3-5 major consecutive incidents away from imploding under the sheer pressure of human perception.

ls cannot merely be slightly better than established programs in giving people an education and a job. You have to outperform by a substantial margin to get everyone on board when you inject yourself into peoples lives to this extent. You ALSO have to convince everyone that you do.

It's pretty unfair and also pretty exciting.

[+] arvindamirtaa|5 years ago|reply
> It’s interesting that pg is a big fan though...

...as an investor.

He may not be intimately familiar with the actual situations of the students from his PoV (mostly throguh conversations with Austen or his team).

But then again, this is far from the first time a negative comment has come out against Lambda. And his position hasn't changed so far. So, there is that.

[+] PragmaticPulp|5 years ago|reply
Investors will only ever speak positively of their portfolio companies.

Lambda school receives high praise in general because so many of us tech types want their value proposition to be true. Just look at how negative the comments about traditional colleges are whenever the topic comes up on HN. Or look at all of the concern about growing student debt loads. Or the concern that 4 year college degrees are not teaching CS students what they need to actually function in real world software jobs.

Lambda School arrived on the scene with a new business model that claims to fix all of that, and they did it as a highly-praised YC startup. It seemed like a home run.

Yet I’ve heard so, so many stories like the OP’s about how they’ve dropped the ball on the educational quality of their programs.

They seem to be approaching this as a typical startup problem where they start with the ugliest MVP that people will pay for and then incrementally improve it only where absolutely necessary to keep the money coming in.

This works if you’re selling a SaaS tool to some company on a monthly basis, but it doesn’t work when you’re taking critical years of young people’s early careers and charging them on the tail end.

Even worse, their customers have an incentive to downplay all of the problems. No one benefits by trashing the source of their education and only career credentials, so instead they say how great it was no matter what. We’re starting to see people pull back the curtain and show what’s really happening.

[+] bambax|5 years ago|reply
> pg is a big fan though

He seems to have a lax definition of the word "earnest". (See his latest blog post.)

[+] dmkolobov|5 years ago|reply
Is it interesting? The fact that pg would ignore the negative behaviors of a YC alumni company is not surprising in the least.
[+] joshxyz|5 years ago|reply
Yep, pg shilling it on twitter is the whackest thing lol.
[+] MivLives|5 years ago|reply
I think that there's another factor to this. The cohorts have grown over time. When you have thirty people in the entire cohort the experience is going to be very different then when you have 130 or more. The earlier students reviews are going to read a lot different then later student ones.
[+] sugarwater|5 years ago|reply
PG should enroll and try the program himself before he offers any praise.
[+] st1x7|5 years ago|reply
People are paying ~$30k for bootcamps? Why not get a degree at that point?
[+] dougSF70|5 years ago|reply
I suspect they set out with noble intentions and then realised it is a conversion-funnel game where they absorb the cost up front with the hope of recouping cost and profit later on. If 100% students passing through landed a job then the unit economics would work provided hard costs are low enough. Business might even be successful at 30% post Lambda School employment. However, as they scale that number will fall because the type of student they recruit might change and the pandemic also having a short term effect. As % post employment falls revenues fall and more students are needed to increase absolute revs. This could also go hand in hand with cost-cutting, of which there is evidence. All in all, an experiment in business model is taking place.
[+] eoinlynch95|5 years ago|reply
Lambda School was the hardest thing I've done. I learned a lot and happy with it but have been having a really hard time finding a job. I am in the EU. Today I spoke to a recruiter who suggested a few changes in my CV. Lambda's staff for helping graduates find jobs are not doing a good job. I've been trying to get endorsed and don't know how many times I had to message different people about getting it sorted.

Overall, Lambda is great and worth doing but its not perfect. Everyone there means well but there needs to be better structure.

[+] ramraj07|5 years ago|reply
Not sure what changed in lambda school, but it's probably not a good year to do any boot camps period.

My alma matter, insight data sciences program, folded last week. Laid off most staff. Grateful for their training, but not surprised that model couldn't survive this year. At the same time I'm hearing offers are starting to skyrocket for some specialists! Let's see how things evolve in 2021!

[+] jamiegreen|5 years ago|reply
Which specialities are you hearing are getting skyrocketing offers if you don't mind sharing?
[+] panorama|5 years ago|reply
It's upsetting to see Lambda making so many negative headlines because it truly is in a position to do the most Good and provide the most value (especially after Dev Bootcamp and Hack Reactor were acquired).

I attended Dev Bootcamp in early 2013, when bootcamps were still largely unknown, and had a transformative time. Just stellar teachers all around that compensated for any perceived weakness in curriculum. The industry business model hasn't changed much, so this implies to me that a bootcamp is really only as good as its instructors and the amount they care.

[+] gexla|5 years ago|reply
Once you're in, why not forget about the quality and problems and just follow the program?

I have only ever gone to low level state schools, but one of my instructors told me that he wasn't there to teach me. He was there to guide me through the curriculum. That stuck.

Stick to the program, sweat out the X months. You'll come out as close to a programmer / designer / X as you would from any other path aside from some sort of paid internship where you can teach yourself on the job.

You could teach yourself. Many do. But most will not.

[+] jo032|5 years ago|reply
If you're self motivated enough to follow a curriculum on your own, you're probably much better off using online resources from MIT/Stanford/Harvard which are not only free but much better quality.

The learning environment described in the linked thread (200 students per teacher, plus blind leading the blind with volunteer mentors) sounds terrible at any price including $0.00. Many MOOCs have community boards, TA's, interactive assignments and cost little or nothing. If their plan is to just bundle that same experience with a job placement sales team and a private slack, then they'll just be yet another for-profit credential mill

[+] Grustaf|5 years ago|reply
Of course you could teach yourself, but why pay 30 k for the pleasure of doing that?
[+] lifteatdev|5 years ago|reply
I paid for the accountability, feedback, and support (which, was provided before I was financially committed). Why would I pay 30k to just teach myself when I can do that with YouTube, Udemy, and documentation?
[+] AzzieElbab|5 years ago|reply
Seems like weird time for moocs and boot camps overall. On one hand they are ideally positioned to teach “real world” it engineering. On the other, by the time teaching materials are polished and presentable the landscape had already moved making your freshly acquired skills outdated. And there is no way to differentiate your offerings if you stick to fundamentals.
[+] crumbshot|5 years ago|reply
> After I was a few Units in (3) and thus financially obligated to pay the ISA, they removed core components of the program.

> Where a mentor was paid to mentor you, you are now forced to mentor other students and vice versa.

So Lambda School is essentially an bait-and-switch pyramid scheme purporting to be an educational institution?

For-profit schools are scams in general, but this one sounds particularly scammy and unpleasant.

I'm not surprised to see it's being backed by YC. Yet another example of venture capitalists using their vast wealth to greedily build even more wealth, no matter who is abused in the process.

[+] kostyal|5 years ago|reply
I went to Lambda School. The cohort based ISA model is groundbreaking. Lambda's execution of it is nowhere near a fraction of its potential