How I Failed to Change Wasmer
mnt.io
mnt.io
That was partly a result of a lot of pressure from larger companies that didn't want to work with Wasmer because Syrus, the founder, was publically rude to other people working on WASM (among other things).
Ivan is a really awesome guy and deserves work that values him. I wish him the best and can vouch for his technical ability.
After this, they wanted another interview. Sure, I guess one 3 hour interview isn't enough. They first scheduled it in the middle of the night my time, but meh time zones are hard. Most egregiously, the interview was to work on the wasmer runtime and fix a bug. Put politely, wasmer is an evolving codebase at a startup. Basically, it's not in the most clean state.
This was a bug involving a virtual file system using WASI. I was fortunate enough to be somewhat equipped to handle this bug. I've sat in on WASI meetings; I remember my file systems; and I've touched a lot of raw WebAssembly. It was still freaking hard as hell. Not to mention, the bug turned out to actually be two bugs sitting next to each other, and partially caused by an unused variable which Rust should warn you about, except for some reason they turned that warning off. Oh yeah and the CEO tried to argue with me about whether Rust warns you for unused variables. Yeah dude, anybody with half a second of experience in Rust can tell you that.
That interview took 2 hours stretching into 3.
After that fun experience, they asked me for yet another interview. I'm sorry but if you're a seed startup hiring anybody with qualifications, you do not get to give me two, 2+ hour long grueling interviews, then ask about a third. You do not get to give me a question that you only solved in the middle of the interview. Your job is to convince me to work for you as much as it is my job to convince you to hire me. Don't waste my time. If you think I'm stupid or lazy or whatever, just reject me.
Maybe it was because I was tired and irritable, but the spirit of the interview did not feel like a collaborative "let's solve this together", but more a "let's see if you're as smart as me". Which just sucks man. I want interviews to be encouraging and emotionally healthy, not just grueling beat downs.
One last petty tidbit, I find it really disingenuous how the CEO markets himself as a "mathematician". I asked about it in the interview and the dude had an undergrad degree in math. Maybe this is my elitism showing but an undergraduate degree is not enough to call yourself a mathematician. There's so much damn work that you need to do to get a PhD. I got pretty damn far in undergraduate math and I still only know a fraction of a fraction of what a PhD knows.
Having the CEO try to dunk on applicants during interviews is not a great sign for any company.
Wasn't the CEO, but the CTO at a startup I was applying with years ago swooped into an interview at a place I was pretty sure I wasn't going to accept an offer from, while I was at the whiteboard solving some stupid thing.
He immediately started arguing with me about some code on the side that was from a previous question, kept interrupting me when I tried to correct him, and then told me I should be cleaner about my work when he finally figured out his mistake.
I thanked the person actually interviewing me (not Mr. CTO) for his time, picked up my stuff and walked out without another word.
I hear a lot of people with a BS in Comp Science that call themselves computer scientists which seems mostly inaccurate too.
It's just odd, these titles and what qualifies you to call yourself a thing.
Mathematician, OTOH, does kinda usually imply that, or at least implies you do something that likely requires beyond undergrad level education.
I have to ask...did you check github after the fact to see if these were real bugs that they used your interview time to solve? Would be funny if there was a commit shortly therafter...
For example, many amateur haskellers have published in academic journals (oftentime in mathematics). I myself have been cited in academic publications despite not having ever gotten a PhD (again related to functional programming). Presumably, someone (And his peer reviewers) thought a blog post I had written about various things in pure FP land was academic enough to qualify for inclusion in a journal.
However, based on my reading of the CEO's profile, blog, etc, it doesn't seem like he fits that definition. In fact, he claims to be a mathematician without as much as one serious blog post or any published content remotely related to mathematics of any kind. I'm not an elitist by any stretch. But to claim to be a mathematician without publishing or speaking on anything related to mathematics, despite being seemingly outspoken on several publishing platforms is incredibly disingenuous
I had no idea that Wasmer, Inc. was a company. The home page, wasmer.io, says nothing about a company and reads like an open source project landing page.
Way at the bottom is a banana button to "Contact Sales" that's just an email address.
What's the product Wasmer is selling?
Clicking through to "about," I see nothing about products, just bios of the team members:
> The CEO, Syrus Akbary, had evidently a lot of pressure on its shoulders.
I suspect that pressure had to do with not having a product for sale in the traditional sense. Maybe the company was viewed as a product, with another company as the customer. If so, it's clear how this could become a pressure cooker.
No product + investor money + team to support = pressure and bad work environment.
Wasmer is an implementation of Wasm.
"Banana" as in the fruit. Monkey as in, well, user. If you want a monkey to follow you, use a banana visible enough that the monkey can see it.
It's basically any giant, clearly visible button.
Good luck to the author, but I would just say one thing if I could meet him: never put so much energy on a job, no matter how much you love the project... it's NEVER worth it. Unless you're one of the real founders, and even then, you're always running the much higher risk of losing everything than of becoming a millionaire.
Go to work. Enjoy it, have fun, but never become emotionally involved with it. If you need social ties, try your local community, working as a volunteer at a school or whatever... at work, you're there to provide your services at a certain agreed time. After that, get out! Go home to your family. Don't let them make you think they're also your family! They're not and they'll forget you as soon as you leave (and you, them). Everyone will be better off by recognizing that.
This may well be true in lots of places; I wouldn't know, having only limited working experience at this point. But I worked for a while at a startup of 5-10 people, and this was very much _not_ the case. Perhaps it was the size of the company. I still have good memories of working there (even if the application domain eventually proved to not suit me well), and even now I have mostly-inactive but still very friendly contact with one of the founders.
There are also good people on this planet.
Its really painful to see people suffer :(
I have driven culture forward before, but a couple lucky breaks can make you think you have more control than you really have and subsequently failing to change other people can feel like a personal failing instead of their issues being the problem. Internalizing that is toxic and that’s what my boss was trying to say but didn’t have the words.
For me, finding that was one of the things that made me pick wasmtime over wasmer for my own project.
Ultimately I went with wasmtime (we ended up using neither at work, too premature) because I just found a frankly disturbing amount of controversy surrounding Wasmer vs. wasmtime, and Wasmer hyping itself up far too much and trying to put down wasmtime.
I'll say as I did 3 months ago[0]. I don't know what's wrong with Wasmer, but there's something _weird_ going on there. I'm not surprised to see this thread.
[0] https://www.reddit.com/r/programming/comments/o1838v/the_web...
If 80% of your engineers are leaving its just a bad workplace. You don't have to apologize for leaving.
The impact of what anybody is building expands beyond the technology itself to all the potential use cases that this new thing enables. It doesn't have a depth of 1. It has a depth of 100.
To imply that "It is just XYZ" in dismissive tone, well that is quite rude and disrespectful at this point in time.
Later in that thread someone surfaced [2] the fact that the CEO had made a sockpuppet account [3] to praise his own projects.
So the OP is not surprising to me. At the same time all of the public communication I've read by the CEO has been polite so I wonder if he is just misunderstood but it's hard to say, it seems like there could also be a lot of bad things happening behind the scenes.
[1]: https://news.ycombinator.com/item?id=24900186
This is a tough story to read for anyone who has been passionate about a startup only to watch it decline under management quarrels. However, I have mixed feelings about this article because the author appears to be a co-founder of the company despite writing much of the article from the perspective of an IC engineer who was taken advantage of. He downplays the role of co-founder later by explaining that he was a “late co-founder” but he also writes about he was responsible for many founder-level activities later in the article.
The only thing worse than startup leadership quarrels is watching one of the founders turn around and try to sink the company by airing all of their dirty laundry on the way out. I don’t necessarily doubt that the environment was toxic, but we’re also only getting one side of the story. If there were any engineers or managers still working at Wasmer and hoping to turn the ship around, this blog post may have destroyed any chance of that. Imagine joining this company only to have a co-founder turn around and advertise to the world how terrible the company is.
I also wish the author would have elaborated more on their original title (“How I failed to change Wasmer”) instead of laying all of the blame on the other co-founder and alluding to toxicities. I suppose the real lesson here is to avoid becoming a co-founder in a company where you know you’re incompatible with the other founders. It never ends well.
What is the alternative though? I mean this developer, at least from their telling of it, had held almost every role in the company up to and including co-founder (at least in title, it sounds like the founder continued to hold him at arms length), if that's all true then what else could this person do other than air out the dirty laundry? They, by their own admission, tried everything they could from within the company and so now they are leaving and letting their truth be known. At worst nothing changes at Wasmer, at best it's a wake-up call that gives the remaining engineers/mangers the opportunity to make the needed changes (if that's even possible).
Simply leaving the company quietly is a huge statement in itself. A founder who leaves an active company and tactfully announces that they’ve chosen to move on speaks volumes.
Airing dirty laundry like this is politics, plain and simple. It doesn’t benefit the author to advertise his issues as a co-founder and it doesn’t benefit the remaining employees who are now at a company known primarily for founder drama.
At the same time not sharing information widely about a toxic member of your community can create a "missing stair"[0] effect where those with inside info know to steer away from the person. Meanwhile, newcomers to the community don't have that knowledge and end up getting burned.
I don't think there's any perfect solution here. We each have to judge for ourselves where to draw the line between public shaming and the duty to protect others from potential harm.
The mature thing to do is to see it for what it is. Not to judge people who've had a hard time.
He was given the title but not a real co-founder position, which frankly was a giant red flag itself.
You are reading too much into the title.
With all respect to the author, I think, from reading this, they gave him the title of co-founder to try to try to stop him from leaving earlier, given 85% of the engineering team had left (it's implied this happend before he became a co-founder). He jokes about it himself, asking "have you ever seen a co-founder on a free-lancer contract?".
If he has the title of co-founder within the company, describes himself as co-founder in his blog, and was responsible for founder-level activities then I think it’s clear that he was, indeed, a cofounder.
I sympathize with his experience, but I also think we need to be careful about letting someone claim the prestige of a cofounder title while disclaiming the responsibilities that come with that title.
Wait, why is that a problem? Let all be aired and let us figure it out for ourselves. We're all adults. If the situation is indeed as the cofounder claims, a cleaned, distilled version of events is untruthful and patronizing.
And if the events aren't as claimed, startups are evaluated on their ability to hire the right people. If the CEO chose a cofounder who would lie and destroy the startup for no good reason, then that also reflects poorly on the CEO.
I, for one, appreciate the signal that Wasmer is a startup to avoid doing business with
People second guess themselves when they write something emotionally charged and it goes semi-viral. So what? Wouldn't you? That doesn't make the article any less valid. A title change is very different from materially changing the actual events discussed, yet you imply something of this magnitude is going on (it isn't).
> The only thing worse than startup leadership quarrels is watching one of the founders turn around and try to sink the company
Maybe for you. For the rest of us, being a toxic founder who exhibits even a third of the bad behaviors mentioned in the article would be grounds for forced removal. Looking over the timeline and events that took place at Wasmer, it's pretty clear this dirty laundry needed to be aired, and everyone (the users, the employees) would benefit if this CEO is actually removed. Perhaps this idea makes you uncomfortable? It's also super disingenuous to paint this as an attempt to sink the company -- clearly it's an attempt to say the product has tons of promise but the CEO needs to go. All other options have been exercised at this point, so going to a public forum with huge industry influence is exactly the kind of thing that could solve this problem that otherwise probably has no solution other than the company floundering. The CEO could quit tomorrow, and then this entire endeavor would obviously be justified and well worth it.
> I also wish the author would have elaborated more on their original title (“How I failed to change Wasmer”) instead of laying all of the blame on the other co-founder and alluding to toxicities
In a literal sense you are right to say that the author doesn't discuss this much, but come on man, it's super obvious -- this "late co-founder" was not given enough control or say in the first place to make any of the changes that needed to be made. Heck, he couldn't even take a sick day without incurring tremendous ire. Most of the article is about that. Did we read different articles? It's like you're going way out of your way to trash this for no real reason other than it makes you feel uncomfortable. Airings of grievances like this are one of the good things about Hacker News and have a positive impact on the industry. I dare you to come up with a single example where that isn't the case. Show me an example of a startup that floundered because of disingenuous HN exposure that wasn't caught until way too late. If anything they'll have a sales bump today.
I do however think that to have that discussion, it would've been much better form to start with the macro level point rather than mentioning his individual family first.
(I know you're not the person I initially replied to)
If the runtime gets embedded in other pieces of software, the impact is a lot more than X stars on Github. You'd need to recursively count all the dependent repos. (Ignoring the fact that you can’t simply measure impact by counting stars).
Kudos to building awesome software and good luck finding your next gig!
> At the time of writing, Wasmer has an incredible growth. In 2.5 years only, the runtime has more than 10’500 stars on Github, and is one of the most popular WebAssembly runtime in the world!