That's a great list but it's really hard to distinguish between fully nomadic (Buffer is the perfect example) and companies that hire many workers remotely. I think most of the listed companies hire remotely (ex. Trello) and have big distributed teams.
I am hugely interested in more examples like Buffer.
You're correct in that the "top 3%" represents the top 3% of the developers who actually applied to Toptal (about 11,000 from all over the world in the last couple months).
However, it's also important to know that the 3% was not arbitrary. We didn’t start by saying, “wow, we should set the acceptance rate at 3%, because that sounds nice” or we somehow knew that the bottom 97% were bad. 3% (and every value you see on our screening process) is simply the result of a numerator and a denominator (successful_applicants over all_applicants). We designed the process to be the bare minimum criteria for us to be extremely confident that we are admitting someone into the network who will thrive here.
The technology stack you are using is not going to affect the chance of getting into the network. Once you have gone through the screening process, you're will be matched by a team of engineers to prospective clients based on your tech stack, availability, time zone and other factors as well.
When it comes to technologies being in demand, it really ranges all across the board: from the standard PHP, Ruby, Javascript etc, over more bleeding edge like React, React Native to niche things like OpenCV are all there.
Full disclosure: I am currently employed as Head of Community for EMEA @ Toptal
Unfortunately, I think this is pretty misleading and incomplete. We do not set anyone's hourly rate, but if you come in expecting $250/hr and there isn't much on your profile to justify that, we will push back to figure out why you think clients will hire you at that rate. Ultimately, your rate is up to you, but it doesn't do anyone any good if you're sitting in Toptal un-hirable. Clients won't hire you if your rate doesn't match your abilities or if there are other people similarly capable but with very different rates from you.
Being spied on? I don't know what that is referring to.
The circumvention clause is real, though we've never even come close to needing to enforce it.
Re refunding all fees: We've never even come close to needing to enforce this, but if someone really botched an engagement, this is a possibility. However, part of the benefit of joining Toptal is that you have the entire network and core team going to bat for you. Good people wouldn't stay long if we were taking all their compensation from them.
I think in both cases the people were essentially lying. (EDIT: probably unintentionally).
I applied to toptal (never did dev work through them, but I would) and after I was accepted, was asked to name my rate - at the time, it was US$60/hr and they said I'd likely find work at that (this is correct, by the way - I did find work at that through them but decided not to take it for unrelated reasons).
In regards to the other comment where bbeneschott used that language, the person was also flat-out wrong. When I applied, step one was a phone call with a person, step two was an automated quiz, step three was a live-coding session over skype with an engineer, step four was an app to build at home, and step five was another skype session with an engineer to review that application. Onerous? maybe, yeah. But definitely not inhuman or automated. Sounds like the other person bbeneschott "called a liar" simply didn't make it through step 2 (I can understand why they'd be upset).
disclaimer: I know bbeneschott personally (through toptal). But I don't earn through toptal anymore and have no financial incentive I can think of to support them.
First, there's no such thing as "unintentional lying".
Secondly, w/r/t/ the person who did their FizzBuzz question in Mathematica: you're trying to suggest that maybe they weren't able to implement FizzBuzz in Mathematica?
Fair point. They were wrong, not lying, and I only used that word because it was the one @idlewords used. Perhaps you're right that I should have corrected @idlewords that accusing people of being wrong is not the same as accusing them of lying. But I love the prose @idlewords writes so I didn't feel that it was my place to correct him on English, when I feel his is better than mine.
I do think it's entirely plausible that an automated coding test could be "failed" by a competent programmer in any programming language, because many automated coding tests are bullshit. It's not Toptal's fault that the state-of-the-art in automated programmer assessment is sub-par. Isn't that the problem Starfighter is solving -- better automated assessment of programmer ability? There's a reason it needs to exist.
> It's not Toptal's fault that the state-of-the-art in automated programmer assessment is sub-par.
Yes it is.
They choose the tools they use, they take responsibility for their results. It is their business, they are responsible for everything that happens on their behalf.
Wait: your CEO replied to this blog post saying that the refund was for billing mistakes. Now, your comment says that it's for if someone really botched an engagement. Which is it?
I'm curious, if you're only doing the top 3%, shouldn't all the work being done be around or above that amount $200+/hr)? I'm probably in the top 3% of devs if we include the mass of people that do any such work. While I don't do hourly, I'd not consider any new projects that don't end up earning at least that much.
Top 3% is sort of a useless metric, isn't it? I'd not be surprised to find that, since most people suck, 97th percentile is a minimum you need to get anything done properly. But within that, there's probably at least an order if magnitude difference in capability (and rates).
Relative perception of others' proficiency for a particular skill is something that subjectively changes as you become better at it. Things you might consider absolutely necessary for a project may not be considered that way for others, and potentially not even recognized by those below your skill level. Projects can ship as a mess of unmaintainable code, after all.
I have close to 25 years of experience; I was told the $100 / hour I was asking for was too much for my region and my rate was set at $50 / hour. I still have an account there, though I didn't yet get any job through them. (So maybe they were right and people thought even $50 / hour was too much?) C# backend programmer.
[I will admit that I blame @patio11 for the $100/hour and "charge more" ideas. /grin]
Oh, edit: I had absolutely no problem with the onboarding process, I think it was just fine. It's only the "no you can't set your own rate" part that I disliked.
Unfortunately, I think this comment is misleading and incomplete. Only one part of the screening process (detailed here: http://www.toptal.com/top-3-percent) is scored automatically, and we still go over ever submission manually to look for cheating and misleading scores (e.g., were you on the right track but just missed something small?).
Our job is to minimize false positives and false negatives. Almost all of us are engineers here. We all know you can't give an automated, standardized test to fully judge anything. The automated test is simply one of several steps, and it does provide helpful information.
Feel free to email me directly at breanden@toptal.com if you'd like more information.
You don't post "projects" on Toptal. Everything we do is time & materials, so the concept of misrepresenting financials based on posted projects simply isn't possible. ARR is calculated by actual sales right now multiplied out for a year of time.
While these technical problems with Upwork’s platform certainly do seem very problematic, I’m not sure that “now is the time to roll out a new freelance platform” is the right conclusion, at least not without some revision.
- For starters, there are plenty of open freelance marketplaces out there that operate pretty much exactly in the same way as Upwork (Freelancer.com, for example). Some are more specialized (like 99designs), but in general the list is pretty endless. Adding another identical platform to the mix is unlikely to accomplish much.
- The real issue is that the operating model for freelance platforms has been broken. This has been the case for a long time, and is unrelated to Upwork’s technical issues.
The open marketplace model falls short in several ways, many of which relate to the “race to the bottom” phenomenon that other responses here mention:
For companies:
a. There’s no quality assurance and profiles/test scores can be fabricated.
b. You have to deal with a large volume of bids, and sifting through them defeats one of the top reasons people hire freelancers: to move fast.
c. There’s very little in the way of protection on any engagement. The risk of your project getting botched by someone incompetent or your freelancer disappearing is significant.
For freelancers:
a. Talented professionals have to deal with absurdly low bids from unqualified people. Many freelancers are forced to lower their rate to below what they’re worth just to get any work.
b. The large volume of bids for most projects means that it can be much tougher to stand out. The most talented freelancers are often the worst self-promoters.
c. There’s no protection from clients with unrealistic expectations.
Long-term, these are the problems that successful freelance models will need to be able to solve at scale. Is this doable? I certainly think so - it’s what we’ve been working on at Toptal. Here’s what we believe the solution looks like, and therefore how we’ve built Toptal to behave:
- All engagements need to solve these problems both for freelancers and for clients. Clients are looking for great freelancers and freelancers are looking for great clients. Therefore the solution is to thoroughly vet both freelancers AND clients.
- Vetting must be rigorous and done in a non-automated, non-riggable, extremely thorough way.
Freelancers must be able to set their own rates and not have to worry about low-bid contests. Only clients with serious projects who are ready to pay freelancers what they’re worth need apply. No race to the bottom.
- There must be a high level of support during ALL engagements, even at scale. Clients should only meet freelancers who were handpicked for their project, and vice versa. This also solves the problem of high bid volumes.
- Support should continue throughout the engagement to resolve conflicts and ensure success.
This is exactly how Toptal has operated for several years now for freelance software developers, and it has worked quite well--even though we made some mistakes along the way. While this has been primarily in the software development space thus far, the principles apply to other types of freelance work as well.
Bottom line: Upwork’s technical issues are certainly problematic, but there are larger issues that freelance platform models will need to solve in the long run.
I wasn't necessarily advocating for another upwork copycat. All I was trying to say by this post is that if there's a new or existing platform that works (unlike Upwork) and is superior in some way (i.e. it solves the problems you pointed out), now is the time to ride the wave.
I was aware of TopTal, but according to your website your platform only solves those problems for 'the top 3%'. Even if we assume that two thirds of the freelancers on Upwork are completely incompetent, then there's still 30% of perfectly competent and capable people who also need work.
Still I earn $40-50/h on Upwork while being from Eastern Europe and TopTal limits me to $25/h because "well, people in your region are poor, you should be poor too".
I want to know if there's a huge difference between charging your laptop and running off of battery sometimes vs leaving it plugged in all the time - I sometimes move to my couch or something to do some work, and maybe leaving the charger behind is actually an additional cost
IMHO, the pain of being handed code is related to its "magic". The more magic, clever, interesting contortions that are present, the less likely you are to get grips on it in any sensible time. Ruby & Rails give lots of tools to create insanely hard to unwind stacks, but I am not sure they are worse than Django & Python (custom loader [importer], decorators everywhere)...
The real bitter pill is that these systems are created with the best of intentions, to make them more maintainable, more DRY, more configurable, more portable... but they generally miss the mark and cause pain for all involved.
I guess that might be true of any inherited codebase in any language - i shudder to think of encountering something written in PERL from an other developer and having to maintain that code.
