I wonder if some escrow protocol would be a solution to the ghosting problem.
Something like, when a company make a job posting they must put some cash in escrow for each candidate they are currently interacting with, and if they ghost you by not responding according to a timely schedule, you get the escrowed money.
Similarly when the job posting has received applications, that have not been rejected in less than a week, they are either autorejected, or they transition to escrow state where they must be handled or else cost money.
It would also have the merit of preventing companies to cast too big a net wasting everybody's time, just to collect data.
Wonder whether it's something that could emerge by people not applying to companies not offering ghost-protection protocol. Or whether name-shaming companies lacking basic courtesy is sufficient.
It could also maybe create more problems that it would solve, because it would add some intermediary in the recruitment process preventing peer to peer contracts. Which brings potentially many problems because it centralize applications for different companies but it also add levers to control bad practices.
Not sure how it could be regulated given international context.
toomuchtodo 1 days ago [-]
I love the direction and attempt. What I think is needed to encourage good behavior is a system that scans company job postings, archives them (Internet Archive), performs analytics as to whether these jobs are being filled (compare to LinkedIn data as best as you can, there will always be leakage), and potentially report non compliance around salary transparency to whomever in each state regulates that. LinkedIn doesn’t care, state regulators aren’t going to do this.
Think in systems. If the desired outcome is more honest and transparent behavior from job posters, the way to get there is public data and analytics to suss out signal of undercurrent intent internally. It’s an observability platform at its core.
Consider partnering with the hiring.cafe folks (search HN for relevant threads) if you want to achieve scale wrt feedback capture from applicants to drive analytics.
dtgmzac 13 hours ago [-]
Thanks for the suggestions! Internet Archive is a really interesting approach. I'd say there are a lot of tools out there that scan job postings and if there were some kind of open standard that everyone used, it would definitely make the job search more transparent.
Schiendelman 1 days ago [-]
I could be mistaken, but doesn't Indeed do this already?
toomuchtodo 23 hours ago [-]
Indeed is an interested party, and their incentives do not align with the general public and jobseekers.
6stringmerc 22 hours ago [-]
I’m doing a long term (read: takes months to collect data) investigative journalism piece regarding hiring practices in the United States that tracks with your musing. As in, I’m using tools to actually document who obtained positions and comparing those visible profiles and credentials with the actual listing put out by companies. The motivation has been through personal experience and seeing…patterns…that I wish to document to the best of my ability in a professional writing project.
I’d be willing to expand my scope if there is interest by any group or entity to support such a project - as in, expand my pool of people actively seeking and applying for jobs (in my case about 10 per week, all within reasonable % of qualification) to explore if data reveals any patterns. It’s a nuanced thing. Basically poking a hole in the “meritocracy” narrative is potentially able to expose other bias at play (age, gender, ethnicity) with a willingness to put such findings into a well researched article / study.
kylecazar 19 hours ago [-]
I was ghosted by a company after going through multiple hours-long interviews. It's the first time it's happened this egregiously.
No communication after 3 rounds, including with very senior people. Last interview was a month ago. Sent two requests for updates. Nothing.
If this helps eradicate the issue I'm for it.
dtgmzac 15 hours ago [-]
That frustration is exactly why I started building this too. I've applied to companies, done their online assessments (OAs), and the status in their portal never got updated, even till now.
The goal is to bring more transparency to the process, so if someone else had received a reply, I could have known sooner and moved on.
hidelooktropic 1 days ago [-]
I would love it if this tool exposes job postings where the company has no intention of actually filling them, just trying to collect resumes. But does it do this?
dtgmzac 15 hours ago [-]
Yup, if you track your job applications there are statuses like "Applied", "Interviewing", "Rejected", "Ghosted", "Offered". So that could fall under "Ghosted". You could also add response dates which is how long it took to receive a reply after applying / interviewing and your interview experience. All these are transparent and everyone can see them.
There's also a "Report Job" button in place for flagging invalid job postings, which you could add a note with the reason for reporting.
However, there is a caveat, since this is community-driven so like another commenter mentioned, there could be bad actors. One way to solve that in the future is to have community moderators / some kind of verification.
It feels unfocused. Application tracking makes sense, but that isn't the same problem as understanding a company's interview process or talking to other applicants about their experience. And frankly, unless you get the entire industry using this for every job application, the community-based features are not going to have meaningful info. Also, the actual problem implied by the title - have I been ghosted - can be answered with static HTML: "If it has been 2 weeks, yes."
So if I were in your shoes, I'd decide what problem I really want to solve, and build an app laser-focused on a solution for that problem. Bring everything else in later, once you know you have PMF for whichever core problem you choose.
dtgmzac 1 days ago [-]
Really appreciate the honest and straightforward feedback. You're right that the community-based features are definitely hard to be useful without a community. It's actually one of my main problems right now, exploring who my target audience is and what the main goal should be before the next iteration.
I actually started this because I wanted to track my own applications but also wanted to solve the problem of making the process more transparent for applicants. Just like the "static HTML" part, a 2 weeks rule is a general guideline, but surely we can do better than that. Especially when there is no recruiters' contact after applying, going beyond a simple yes/no answer by providing community-driven data (historic and current) would help a lot (as seen in the links I cited). Thanks again for the insights!
pogue 1 days ago [-]
I wonder if you could implement some features used by the advertising industry and when you submit an application, include a tracking pixel/web beacon that would trigger when your application was viewed and returned some analytics like if it was viewed in an email client or web browser. You would then be able to see what's going on with the other end of your job application.
dtgmzac 15 hours ago [-]
Interesting idea! I'd love to know if that's possible too.
When we submit our application, usually we also submit a resume. Personally, I do add UTM code (e.g. utm_source) to my links in my resume so I know if it was visited through analytic tools like Mixpanel. However, that's assuming that the person viewing the resume clicks on it.
pogue 14 hours ago [-]
What format do you submit your resume in? In a doc or PDF you should theoretically be able to load an externally accessible beacon. Depending on how the person is viewing it, you could get some interesting results, including if an LLM looks at it first.
dtgmzac 13 hours ago [-]
PDF format.
> including if an LLM looks at it first.
Wow, do you have any links / examples of this kind of tracking or have you seen that done anywhere? If it is effective, I think it would be really useful!
trod1234 1 days ago [-]
The issue with community-based/crowdsourced features as well is that you inevitably have some bad actors that join up to poison your features.
If the noise floor threshold is too high, the signal you are looking for can't be differentiated.
dtgmzac 15 hours ago [-]
Great point. A moderation system or verification features could definitely help reduce the risk of fake accounts or bad actors. If this gains enough users, it’s definitely something I'll have to tackle. Appreciate the feedback!
trod1234 1 days ago [-]
I disagree, in the past I've seen positions where they may sit on applications for a month or more. One of my last jobs actually ended up being one of these, and we went through 9 different interviews (over 2 1/2 months after that) before I was hired at the time.
Its not so cut and dry.
demarq 1 days ago [-]
This has to be Microsoft!
trod1234 21 hours ago [-]
A lot of large companies do this, especially in the biopharma sector.
vivzkestrel 18 hours ago [-]
is that shadcdn or tailwind css? looks really nice
dtgmzac 15 hours ago [-]
Thanks! It's HeroUI (previously NextUI), which is a UI library that is built on top of tailwind css.
You can check out the GitHub repo too if you're interested!
trod1234 1 days ago [-]
Interesting, I was thinking about doing something similar just the other day given all the shennanigans going on with ghost jobs.
There's a lot of time wasted individually by candidates vetting companies. For instance I've had to resort to using OSINT techniques and public records to weed out fake jobs because my cold interview->conversion rate is abyssal (1:1000+).
I've seen companies and positions being advertised where when you look up the secretary of state info, its either been shut down/suspended, or the directors/owners listed are dead people (i.e. semi-recent obituaries +- 1 year). The positions posted are always newer than the deceased person. I've also seen a lot of business info posted resolving to a UPS store address. Definitely presents as red flags when looking.
It would be useful to be able to crowdsource a lot of the due dilligence that must be done individually that could easily be automated as well as see the average turnaround time before rejection, and identify job postings that continuously roll (identical and reposted every 3-6 months)
dtgmzac 13 hours ago [-]
> There's a lot of time wasted individually by candidates vetting companies. For instance I've had to resort to using OSINT techniques and public records to weed out fake jobs because my cold interview->conversion rate is abyssal (1:1000+).
Do you automate the whole process? That's really a lot of applications.
> It would be useful to be able to crowdsource a lot of the due dilligence that must be done individually that could easily be automated as well as see the average turnaround time before rejection, and identify job postings that continuously roll (identical and reposted every 3-6 months)
Currently the platform does this! On the company page, you can see job postings split into "Opened" and "Closed" statuses, as well as when people applied and the turnaround time.
I'm also looking into automated approaches like a browser extension which could be useful if more people uses it.
pogue 1 days ago [-]
Can you share some of the techniques you used? It might be helpful to get some simple scripts up to check if the address listed is a UPS store or registered agent or something and create a browser extension for job search sites that would automatically hide listings on job search sites that meet those criteria.
trod1234 20 hours ago [-]
I use a wide range of techniques, quite a few are covered in Michael Bazzel's book on OSINT and its a large amount of material. The book is well worth getting if you haven't had exposure to these things.
I'm basically looking for red flags, so inconsistency in website information vs. business and government records. This includes Secretary of State information which may need fuzzy matching, but includes Agent of Service, Address, and Company Leadership, as well as renewal dates, business status, renewal status. Legitimate businesses are not inconsistent here.
I look at the website for how they get customers, and I verify their contact information is functional (usually through a VOIP number). A business without a phone line is usually not a business. Having a listed phone in the reassigned numbers database is a pretty big red flag as well.
I'll look for social media business accounts, county court records, coroner reports, etc. blackbookonline.info and judyrecords.com have been my goto for awhile. The latter search engine was broken for awhile though because they provided too much info and were sued. Aside from that sometimes I get hits on OpenPayrolls.
Companies that hide salary/company name, and postings that were posted without disclosure by recruiters are pretty big red flags as well. I'm sure there's plenty else but that's what I've been doing.
pogue 19 hours ago [-]
Very helpful, thanks. I came across this site from the FCC for reassigned numbers, but I'm not quite sure how to use it. Is this the right site? www.reassigned.us
Was the Michael Bazzell book you're referring to
"Open Source Intelligence Techniques: Resources for Searching and Analyzing Online Information" or the "OSINT Techniques: Resources for Uncovering Online Information" (both around $44 on Amazon)?
trod1234 14 hours ago [-]
There are a number of services that allow you to query that database. I'm not familiar with reassigned.us
> OSINT Techniques...
Yes that's the one. Its in its 10th edition now I believe, pubbed in 2023.
ISBN 9798366360401
pogue 13 hours ago [-]
Do you have any suggestions for a free tool to query the reassigned numbers database?
Thanks for the help!
trod1234 1 hours ago [-]
Unfortunately I'm not aware of any free tool that lets you query it. Each query costs me a few cents so I try to minimize queries. Twilio has a package available in their API to do this.
Something like, when a company make a job posting they must put some cash in escrow for each candidate they are currently interacting with, and if they ghost you by not responding according to a timely schedule, you get the escrowed money.
Similarly when the job posting has received applications, that have not been rejected in less than a week, they are either autorejected, or they transition to escrow state where they must be handled or else cost money.
It would also have the merit of preventing companies to cast too big a net wasting everybody's time, just to collect data.
Wonder whether it's something that could emerge by people not applying to companies not offering ghost-protection protocol. Or whether name-shaming companies lacking basic courtesy is sufficient.
It could also maybe create more problems that it would solve, because it would add some intermediary in the recruitment process preventing peer to peer contracts. Which brings potentially many problems because it centralize applications for different companies but it also add levers to control bad practices.
Not sure how it could be regulated given international context.
Think in systems. If the desired outcome is more honest and transparent behavior from job posters, the way to get there is public data and analytics to suss out signal of undercurrent intent internally. It’s an observability platform at its core.
Consider partnering with the hiring.cafe folks (search HN for relevant threads) if you want to achieve scale wrt feedback capture from applicants to drive analytics.
I’d be willing to expand my scope if there is interest by any group or entity to support such a project - as in, expand my pool of people actively seeking and applying for jobs (in my case about 10 per week, all within reasonable % of qualification) to explore if data reveals any patterns. It’s a nuanced thing. Basically poking a hole in the “meritocracy” narrative is potentially able to expose other bias at play (age, gender, ethnicity) with a willingness to put such findings into a well researched article / study.
No communication after 3 rounds, including with very senior people. Last interview was a month ago. Sent two requests for updates. Nothing.
If this helps eradicate the issue I'm for it.
The goal is to bring more transparency to the process, so if someone else had received a reply, I could have known sooner and moved on.
There's also a "Report Job" button in place for flagging invalid job postings, which you could add a note with the reason for reporting.
However, there is a caveat, since this is community-driven so like another commenter mentioned, there could be bad actors. One way to solve that in the future is to have community moderators / some kind of verification.
Oct 2023, 130 comments: https://news.ycombinator.com/item?id=37792507
So if I were in your shoes, I'd decide what problem I really want to solve, and build an app laser-focused on a solution for that problem. Bring everything else in later, once you know you have PMF for whichever core problem you choose.
I actually started this because I wanted to track my own applications but also wanted to solve the problem of making the process more transparent for applicants. Just like the "static HTML" part, a 2 weeks rule is a general guideline, but surely we can do better than that. Especially when there is no recruiters' contact after applying, going beyond a simple yes/no answer by providing community-driven data (historic and current) would help a lot (as seen in the links I cited). Thanks again for the insights!
When we submit our application, usually we also submit a resume. Personally, I do add UTM code (e.g. utm_source) to my links in my resume so I know if it was visited through analytic tools like Mixpanel. However, that's assuming that the person viewing the resume clicks on it.
> including if an LLM looks at it first.
Wow, do you have any links / examples of this kind of tracking or have you seen that done anywhere? If it is effective, I think it would be really useful!
If the noise floor threshold is too high, the signal you are looking for can't be differentiated.
Its not so cut and dry.
You can check out the GitHub repo too if you're interested!
There's a lot of time wasted individually by candidates vetting companies. For instance I've had to resort to using OSINT techniques and public records to weed out fake jobs because my cold interview->conversion rate is abyssal (1:1000+).
I've seen companies and positions being advertised where when you look up the secretary of state info, its either been shut down/suspended, or the directors/owners listed are dead people (i.e. semi-recent obituaries +- 1 year). The positions posted are always newer than the deceased person. I've also seen a lot of business info posted resolving to a UPS store address. Definitely presents as red flags when looking.
It would be useful to be able to crowdsource a lot of the due dilligence that must be done individually that could easily be automated as well as see the average turnaround time before rejection, and identify job postings that continuously roll (identical and reposted every 3-6 months)
Do you automate the whole process? That's really a lot of applications.
> It would be useful to be able to crowdsource a lot of the due dilligence that must be done individually that could easily be automated as well as see the average turnaround time before rejection, and identify job postings that continuously roll (identical and reposted every 3-6 months)
Currently the platform does this! On the company page, you can see job postings split into "Opened" and "Closed" statuses, as well as when people applied and the turnaround time.
I'm also looking into automated approaches like a browser extension which could be useful if more people uses it.
I'm basically looking for red flags, so inconsistency in website information vs. business and government records. This includes Secretary of State information which may need fuzzy matching, but includes Agent of Service, Address, and Company Leadership, as well as renewal dates, business status, renewal status. Legitimate businesses are not inconsistent here.
I look at the website for how they get customers, and I verify their contact information is functional (usually through a VOIP number). A business without a phone line is usually not a business. Having a listed phone in the reassigned numbers database is a pretty big red flag as well.
I'll look for social media business accounts, county court records, coroner reports, etc. blackbookonline.info and judyrecords.com have been my goto for awhile. The latter search engine was broken for awhile though because they provided too much info and were sued. Aside from that sometimes I get hits on OpenPayrolls.
Companies that hide salary/company name, and postings that were posted without disclosure by recruiters are pretty big red flags as well. I'm sure there's plenty else but that's what I've been doing.
Was the Michael Bazzell book you're referring to "Open Source Intelligence Techniques: Resources for Searching and Analyzing Online Information" or the "OSINT Techniques: Resources for Uncovering Online Information" (both around $44 on Amazon)?
> OSINT Techniques...
Yes that's the one. Its in its 10th edition now I believe, pubbed in 2023. ISBN 9798366360401
Thanks for the help!
https://www.twilio.com/docs/lookup/v2-api/reassigned-number