Show HN: CakeResume – Drag and drop resume snippets to build a unique resume
cakeresume.com
cakeresume.com
Maybe (?) that works for the super-elite echelon who don't really have to "job hunt", and who maintain a resume only as a formality. But for the other 95%+ of plebs who have to actively apply or work with recruiters... MS Word is the only game in town, like it or not.
I loved the idea behind JSON Resume (https://jsonresume.org) a year ago (i.e. separate the content from the styling, store the content in source control and track it or branch as needed, and apply the content to whichever template you like). But I didn't like the limited export capability... so I wrote Resume Fodder (https://resumefodder.com) as an open source alternative. I'd love to create more template options when I have time... pull requests welcomed!
The people behind these resume generators know that MS Word is what most people need, but choose not to support it because that would undercut their business model. The business model dream is for you to host your content on their server, so they can mine it and sell access and hopefully cut off a slice of LinkedIn's pie.
So be it, but I just don't think there is a viable business model there. LinkedIn is already LinkedIn... and I suspect that smaller rivals will run into the same wall as Diaspora, Ello, and all the other would-be Facebook replacements. I think that people who check out these resume generators REALLY are looking for... a resume generator! And if it doesn't let you save your resume in the format that most people demand, then it isn't particularly useful in the real world.
That's stretching it. Most companies are totally fine with PDF these days, not in the least because it is WYSIWYG for screen and print and just works. Looks a lot more professional too; especially if you are looking for a job in IT.
When recruiters or temp agencies ask for Word documents, they want this because then they can strip out all your personal data and pass the documents on to the companies (their customers) looking for employees. With PDF they need to do more work.
This is correct. Agency employees are taught to remove contact info, apply their own header, and trim the length (5 page maximum) before sending to clients.
If only some PDFs will work for them, they wouldn't accept them at all.
[0] http://tex.stackexchange.com/questions/297426/spacing-betwee...
Places I'd actually want to work are far morely likely to accept or even require PDF.
Better to just have an explicit conversation with the recruiter agreeing to your ethical/personal boundaries. Based on their response you'll probably get a good clue whether or not they respect that.
There are ethical recruiters but as with any group of people you have to sort out the good ones.
(TEKsystems in Southfield, MI, for what it's worth. But this was also 1999-ish.)
I've since found out that it's a pretty common practice and usually the candidates play along just to get the job.
For example, I've had recruiters add in stuff like "2 yrs exp in PHP" when the closest I had gotten to PHP then was editing Wordpress to make it work like it should.
Other time they'd edit my contact info out, change wording, stuff like that. I'm not happy with edits being made without my approval, so I stopped giving them an editable format.
I never saw anything super egregious like lying about skills - but that's not to say it never happened.
Exporting to Word would be simple change if merited, so that is really more of a feature/strategy suggestion rather than evidence the business is fundamentally not viable.
Regarding competing with LinkedIn, it hard to judge without knowing the full vision. OP said this was step 1 towards a platform. Besides LinkedIn does lots of things suboptimally, and other people have built businesses that work along side their ecosystem.
Finally other comments are correct when they say recruiters only want Word to strip your info. Couple fyis on this point:
1) Stripping contact info is not unethical. If a recruiter worked hard to find you and/or build a relationship they should not have to worry about companies using their work to cut them out of a deal.
2) Adding in skills without your knowledge is way unethical. Ask up front if they do this. Some also scrape your resume (and tons more) and submit without you knowing, then if a company is interested they call you and say XYZ company is interested in you. Most people don't complain about this part because its a free job offer, however personally I don't like and always ask what their source was.
3) Having only PDF and not Word is very unlikely to hurt you. If recruiters are interested they will just OCR into Word in about 30 seconds.
And our customers are mostly entrenched industrial suppliers and government-owned private entities with byzantine processes, not some tech startups.
In my experience they love to use Office internally, but not for external documents.
I feel once editing is done, PDFs are much more professional because they open without any virus warnings, update notifications, or "this document is in read-only mode" banners.
That said, yes, when editing is NOT done a PDF is almost impossible to deal with.
Have you carried out any research/surveys for Resume Fodder to find out what percentage of people are encountering this?
Edit: I'm in Ireland, so maybe this is a US thing, but I've always thought we were a bit behind here in terms of trends of software usage.
As for output formats, I was planning to integrate HackMyResume (http://please.hackmyresume.com/). Then using the library of existing templates would have been possible. The graphic design quality of some of those templates isn't what I'd want though... I had ideas about enabling HTML templates to support interactivity in some fashion.
But HackMyResume is Node based... and Node is annoying as I found out. I see you are using Go, so no idea what it might take to integrate Node or reuse the logic. Why can't you use the JSONresume templates available?
Then I decided to stop being a developer :) Life is ... different now. I'd recommend it if one is so inclined.
I was thrashing about, so I decided to go to grad school to get an official certificate aka MSc in machine learning. The statistics part is surprisingly interesting, it's technical so I'll make some money, but it's not the grind of development. I would not take a data engineering position, for example. My last job involved some work in this area and it was a breath of fresh air re: my desire to deal with computers. I wish I'd looked into it earlier. My engineering/dev background + this will hopefully make HR/recruiters think I'm a unicorn. A highly paid unicorn.
I have this book, you'd probably enjoy it:
http://www.pobronson.com/index_what_should_I_do_with_my_life...
But JSON Resume is entirely open source already: https://github.com/jsonresume
Why not simple add a Word output to the existing project? You also created a new schema, which is almost the same, but not quite.
I don't understand the purpose of re-inventing the wheel here. The other project has dozens of contributors, I'm going to assume it stays up to date more.
--- Edit: I see your product does actually integrate with theirs. Still seems like it could have all been one product, though. I don't understand what you mean by "an open source alternative."
For example, you can't add projects to a job; you can't assign skills to a project or a job. For people who did consulting, this is a must, but not available. There are many other issues, but those are, for me at least, the road blocks.
On my last job hunt, nearly every employer and job board accepted resumes in PDF.
For the most part, the only people who wanted Word were recruiters, and I'm done doing business with them. Every single job I've ever gotten has been the result of me directly applying, and all recruiters have ever done is waste my time and spam me.
I agree that this is still kind of an open problem, but unlikely to be a business opportunity in the making.
I wrote https://github.com/awalGarg/cv-maker/ which generates the resume in markdown. From there, you can convert it with pandoc to basically whatever you want. It is not much polished yet, though.
I love json resume but last time I couldn't even find a good way to generate pdf's just html, much less doc.
standard template doesn't seem to be working.
http://please.hackmyresume.com/
But it is a Node app.
Developer Story does have a PDF export but it's very limited to including everything which is a bit heavy, and no UI customization right now.
I'd be more likely to explore a tool like CakeResume if they imported from an SO CV i.e., if the cost of adoption was reduced. I think this reflects the typical developer opinion on most resume generators.
I'm not saying SO has to be the standard, it could be a JSON schema with even more features, but we really need some standard.
That's the approach i'm taking with JobRudder [1]. You enter your "data" and we generate resumes, achievement reports, timelines etc. We're looking into updating LinkedIn, posting to twitter (for those who like that) and other integrations.
I'm happy more folks are thinking this way.
I've tried many varying levels of information (just a few weeks ago we had multiple large screenshots and more text) and the feedback is always a bit mixed (between those who think it's cluttered/dense and those who want more). It's not right yet, but we're working on it. Maybe we'll switch to a video.
In the meantime, feel free to try it out (30 days free). No pressure.
EDIT: One other thing. What do you think would make it easier to get value immediately? Starting with a resume you have now? LinkedIn?
Thanks again.
I wish JobRudder could help your use case. I don't know enough about getting ahead (what it takes to get more contracts, higher fees etc) as an independent contractor, but i'm very much willing to learn if you'd like to talk about it.
Thanks again.
The challenges I've experienced relate more to the business and administrative functions and in general optimizing the amount of time spent on non-billable things. One analogous challenge a contractor faces is figuring out how to make the jump from software contracting into technical consulting.
That's wonderful situation to be in :)
> The challenges I've experienced relate more to the business and administrative functions and in general optimizing the amount of time spent on non-billable things.
This is a tough one and it's something i struggle with too. I take it you already have the usual suite of apps to help (xero/freshbooks, timely/toggl, hellosign etc).
> One analogous challenge a contractor faces is figuring out how to make the jump from software contracting into technical consulting.
Also tough, but more familiar. This jump always feels like a matter of specialization to me. It usually means leaving the variety of being a generalist behind. Luckily there are many ways to specialize: Functional. Technical. Process Engineering. You can even combine these specialties.
The less technical specialties require more nuanced proof of your abilities. It's definitely a good idea to keep notes of applicable project, moments, achievements.
I was a consultant for 7+ years (taxes + tech), although not an independent one, and i'll say this: it takes even more non-billable time to do it right.
Do you have special skill set?
2 - Exactly what you said.
3 - Not necessarily. I've done a ton of work with Python and a good amount using Python for APIs and web apps. I'm doubling down on that currently, but that's probably too general to be considered a speciality. It's definitely something I'm trying to feel out.
Good on you. Knowing where where you want to go is a large part of the battle.
Developer Story is a really good product but I haven't yet taken the time to create mine. I should go do it now.
Otherwise looks nice, atleast in the images.
Getting a "Username is too short (minimum is 2 characters)" after signing up with Twitter.
Maybe you got something useful to offer, but you need to demonstrate value first before hassling me with that.
In our job application frontend we don't even require logins or passwords. Only after you have finished your application and confirmed the privacy agreement we send an email with a confirmation link, only then you can add a password to follow the progress of the application, but even that is optional.
I noticed smartrecruiters.com follow this format and it was such a pleasant experience using it.
On the topic of resume building though, while I haven't had the need for making one in a long time I have this idea I just want to throw out there.
Just make like an HTML page with every possible detail about your history. Then, in e.g. Chrome, you can easily hide the different sections/words that isn't related to the specific job you're applying for. Then print as PDF. Seems like it would be easy to maintain and "generate" the resumes per application.
I do this in LaTeX. I have the full document and just comment out in the code what I don't in each case before compiling the PDF. As it is a plain text file I have it in a git repo so it's trivial to look for older versions or attach tags to specific CV versions I've submitted.
Looks nice from the landing page, but maybe a bit image-
heavy for a resume isn't it?
Just make like an HTML page with every possible detail
about your history. Then, in e.g. Chrome, you can
easily hide the different sections/words that isn't
related to the specific job you're applying for.
It strikes me as really hard to do that programmatically without a mess of show/hide logic that defeats the point.
First, you could simply require the registration before allowing them to export so you still capture the data.
Secondly, I'm afraid the usability and slickness are so nice that you will lose more than gain by not letting people get a quick chance to see how good it feels.
Glad I got that off my chest. My opinion and $1.85 will buy you a cup of coffee at Starbucks, so fwiw.
You can save it as XML if you want to update it later. You can email it as PDF or .docx file. And it supports many languages.
I personally don't have it on my resume because I feel that it's un-necessary and a bit unprofessional. But I can see it's appeal (there have been too many instances where there's one person who does the phone interview and another the in person interview)
How does having a photo on a resume help prevent this? Are you referring to the interviewer? (i.e one person who conducts the phone interview and another the in person interview)
But what I mean is, resumes get passed around (from HR to team lead etc.) and having a photo helps identify the person being interviewed. (generally one person does the first interview, followed by an technical interview done by someone different - and it's very easy to get different people answer these interviews till the very last in-person interview - infact, I've seen a couple of friends do that at the big 4.)
I'm not saying that this will help but it's gives everyone a false sense of security.
eg Germany: http://www.thelocal.de/jobs/article/47649
It's definitely considered strange in most of the USA.
You should probably do a better job of calling out the things people expect to put on a resume, instead of hiding them under generic non-resume terms.
Farsighted Note: I am aware of PDF export feature, but its very ugly..
I like the idea. Your site looks great - clean and attractive. The animations are enticing.
I've spent a lot of time exploring and building career tools [1], and my observation so far has been that "formatting and style" isn't the main problem with resumes. Seem to me that most people struggle with "what" to put in a resume and "how" (phrasing, concreteness etc) to put it.
Good luck and congrats on launching.
Also, because my brain has been damaged by modern programming languages, it seems like uniqueness should be defined as (please excuse my rusty logic syntax):
R = resumes, r \in R, unique(r) -> (\forall s \in R-{r}, r != s) (for some definition of equality).
A collection of unique values constitutes a set with exactly one element.
My website is generated with a static website builder, hugo, and uses wkhtml to generate an optional pdf. It just pulls a subset of what I want to display from the exhaustive, version controlled json file. My entire website pulls contact information, name, location, social media profiles etc information from that one json file. It separates content and presentation and I keep it all in one file. If I wanted two versions of the information, I just change the html document and what I pull.
I usually just whip up a good looking HTML page and then use wkhtmltopdf :p