If your LinkedIn Easy Apply drawer is full of silence, one comment buried in the September 2026 "Ask HN: Who is hiring?" thread is worth reorganizing your week around. A repeat poster disclosed 216 applications and roughly 10 offers from a single August posting - a 4.6% offer rate on a channel most people still treat as a curiosity.
The 216-to-10 disclosure, in context
Catalyst·Wayfare AI, a remote agentic-coding employer, received 216 applications on their August 2026 HN posting and issued offers to approximately 10 candidates. The disclosure appeared in the September 1, 2026 thread that has since drawn 254 points and about 397 comments in its first two weeks.
The number matters because almost every published benchmark for online applications measures interviews, not offers. LinkedIn Easy Apply lands in the 2 to 8% range for interviews. A full, tailored LinkedIn application does 10 to 20%. HN's 4.6% is offers. Convert the LinkedIn interview rate to offers at the usual one-in-four to one-in-six rule of thumb, and HN is running 5 to 10 times better on the outcome that pays rent.
216 applications, ~10 offers, self-disclosed by the poster in the September 2026 "Who is Hiring" thread.
The reason is not that HN is magic. The inbox at the other end is a founder's, not a recruiter's queue, and the applicants who bothered to comment on a text-only thread had to read the whole post before they could reply. The Catalyst·Wayfare poster also wrote that they intend to have a personal relationship with every hire - a signal that this is a founder-read inbox, not a coordinator's shortlist.
Why HN converts when Workday does not
The direct-apply channel works because there is no parser in the middle and no recruiter triaging 300 resumes for a coordinator. Consider the load on the other side. Workday Recruiting customers processed 173 million job applications in the first half of 2024, up 31% year over year, while requisitions grew only 7% to 19 million. Applications grew about 4x faster than openings. 97.8% of the Fortune 500 runs an ATS, and the average corporate posting collects 300+ applications per hire. The first 10 to 30 applications on a Workday req get a real human read; the rest are ranked by keyword score.
HN inverts every part of that stack:
- No parser. The founder reads the email body. No PDF machine in the middle.
- Self-filtered pool. Applicants had to scroll a text thread and read a paragraph. Auto-apply bots do not touch HN.
- Explicit instructions. Most posts ask for a specific artifact ("name one AI feature you shipped with real users"). Ignoring it culls you.
- Small field. 216 applications for one role is comparable to a Fortune 1000 posting's first 48 hours (250+), but with zero bot traffic.
- Founder answers own email. The Catalyst·Wayfare poster committed to personal relationships with every hire.
The mechanism is boring and it is the whole point: when the reader is the hiring manager and the founder, personalization stops being polite and starts being decisive.
The funnel comparison, side by side
Direct benchmarks between HN and LinkedIn are rare because the two channels report different stages of the funnel. Here is the honest comparison, keeping the source stage attached to each number.
| Channel | Applications | Interview or offer rate | Source |
|---|---|---|---|
| HN "Who is Hiring", Catalyst·Wayfare, Aug 2026 | 216 | ~4.6% offer rate | September 2026 HN thread |
| LinkedIn Easy Apply (broad) | n/a | 2 to 8% interview rate | appycan.com |
| LinkedIn Full Application (tailored) | n/a | 10 to 20% interview rate | appycan.com |
| US tech median, 14-day window | n/a | 5 to 7% recruiter response | jobaholic.app |
| US tech top decile | n/a | 12 to 15% recruiter response | jobaholic.app |
| Fortune 1000 posting, first 48 hours | 250+ | first 10 to 30 read in depth | jobaholic.app |
Two things fall out. First, HN's application volume per role is in the same order of magnitude as a Fortune 1000 posting's first two days, but the pool is pre-filtered by self-selection. Second, on outcomes, HN's offer rate compares to LinkedIn's interview rate at parity or better, which means HN wins the actual conversion by a factor of five to ten.
HN volume is at eight-year lows, and that helps you
Fewer roles per thread is good for applicants, not bad. A recent "Who is hiring?" thread hit an eight-year low, back to January 2015 posting levels, per hntrends.com. Each posting draws a smaller share of the thread's readers, and those readers are still self-selected humans rather than auto-appliers.
AI mentions in HN postings have nearly doubled since October 2022, from about 10% to close to 20% of posts, so the roles that remain are concentrated where founder cash and technical hunger overlap: applied AI, infra, agentic tooling, early-stage product engineering. For an engineer with a real project shipped in the last twelve months, that concentration is a tailwind. There are fewer HN roles than there were in 2021, but the ones that exist read your paragraph.
The resume rules invert when a founder is reading
A resume aimed at a founder inbox looks nothing like a resume aimed at Workday. Where the ATS resume optimizes for parser tolerance and keyword density, the HN resume optimizes for the founder's first ten seconds of skimming your email body before they ever open the PDF.
The old rules, and what replaces them:
- Keyword density → specific artifacts. Do not spray "distributed systems, Kubernetes, Rust." Name the service, the traffic, the outcome. Catalyst's post asked for one AI feature shipped to real users. Answer that in line one.
- Reverse-chronological wall → the paragraph that matches the post. Move the two bullets that map to the posting to the top of the resume, above the header line if you have to.
- PDF as primary artifact → email body as primary artifact. Put the answer to the post's question in the first two lines of the email. Attach the PDF for depth.
- Neutral, formal tone → the tone the post is written in. HN posters write casually. Reply in kind.
- Cover letter as ritual → cover letter as filter. If the post has a question, answering it is the cover letter. Skip the "I am writing to express interest" opener.
The AI-generated-resume tax matters most here. 49% of US hiring managers report auto-dismissing resumes they suspect are AI-written, and 62% reject AI resumes that lack personalization. A founder can smell a generic LLM cover letter in one sentence; a Workday parser cannot. The trick is not to avoid tools, it is to use them for the boring reformatting and keep the specifics human.
That is the exact split Refolk is built for: paste the HN comment as the job description, and Refolk rewrites your resume from your own history against that specific post, drafts the first cover paragraph in the tone of the ad, and scores how well your background actually matches what the founder asked for. You still write the sentence about the AI feature you shipped. Refolk handles the surrounding rewrite so you can send ten tailored applications the day the thread goes live instead of one.
On HN, the cover letter is not a ritual. It is the filter.
The startup engineer pool is smaller than you think
The direct-to-founder channel is uncrowded relative to LinkedIn, and Refolk's index puts a number on it: 791 US software engineer profiles actively signal "startup" in their headline or current work. That is the shallow but real pool of engineers fishing where HN posts.
In a sample of the top current employers among that pool, the companies that appear repeatedly are exactly the size profile that posts on HN rather than running a Workday funnel:
- Redo
- Brook Health
- Mudflap
- Citra Space
- Raptor Maps
- Bluebird Kids Health
- Privy
- Maior
- DoorDash
- Cultivo
Six of the top 25 startup-signaling profiles concentrate in the San Francisco Bay Area, but the roles themselves are overwhelmingly remote. The gap between engineers who could plausibly work at these companies (millions) and engineers who are actively signaling they want to (under a thousand) is where the offer rate lives.
From Refolk's index of professional profiles, the actively-signaling pool fishing in HN's direct channel.
A playbook for the September 2026 thread
Treat the thread as a channel with its own conversion mechanics. Here is the sequence that maps to the 216-to-10 disclosure.
- Read the thread on day one. The founder is refreshing their inbox in the first 24 hours before selection fatigue sets in. Same leverage as the Fortune 1000 "first 10 to 30" rule, different clock.
- Filter with an indexer, not the raw thread. The whoishiring account itself points readers to hnhiring.com, hnjobs.emilburzo.com, nchelluri.github.io/hnjobs, and nthesis.ai/public/hn-who-is-hiring. HN Match Maker (hnmatchmaker.com) also cross-references "Who Wants to be Hired?" posts.
- Pick 10 posts, not 40. The channel rewards specificity. Auto-apply patterns of 30 to 50 applications per day are the opposite of what works here.
- Answer the question in the first two lines of the email. If the post asks for a specific artifact, that is the entire filter.
- Attach a resume tailored to that post. Not your master resume. The version where the two bullets closest to the post's stack sit at the top.
- Skip the ATS section of your day. Half of your budget for tailored HN replies, half for high-effort direct outreach. Nothing left for Easy Apply.
What this changes about your week
Nothing about HN's 4.6% offer rate replaces a broad job search. It reframes where the marginal hour goes. If you are sending 50 Easy Apply applications a week and landing three replies, moving 15 of those hours into 10 hand-written HN replies is likely to double your interview count and multiply your offer count by more than that. The channel is small, the field is self-selected, and the resume rules invert. If you write for the founder reading the email, the funnel rewards you.
FAQ
Is Hacker News's "Who is Hiring" really better than LinkedIn?
For engineers with a shippable artifact from the last twelve months, yes, on outcomes. The 4.6% offer rate from one August 2026 posting is 5 to 10x better than LinkedIn Easy Apply once you normalize interview rates to offer rates. It is not better in volume. HN threads have been shrinking to eight-year lows, so use it as a concentrated channel alongside broader search, not as a replacement.
What should the first line of my HN reply say?
Whatever specific question the post asked, answered concretely. Catalyst·Wayfare asked for one AI feature shipped to real users and one discovery-led engagement. The winning replies named the feature, the users, and the outcome in the first sentence. If the post did not ask a question, name the closest project you have shipped to the stack or domain the post described, with a number attached.
Does a founder inbox care about ATS-safe formatting?
No. There is no parser between you and the founder, so one-column, keyword-dense, image-free formatting buys nothing. What matters is that the body of your email answers the post's question in the first two lines and the resume opens with the two bullets closest to the role. Save the ATS-safe version for Workday postings.
Can I use an AI tool to write my HN reply?
Yes, for the rewrite; no, for the specifics. 49% of hiring managers auto-dismiss resumes they suspect are AI-written and 62% reject AI resumes without personalization, and founders on HN spot generic LLM tone faster than corporate recruiters do. Use a tool like Refolk to tailor the resume against the specific post and draft the surrounding paragraphs, then write the sentence about your actual shipped work yourself.