How talenthunt works

talenthunt turns a CV into a conversation. Instead of asking someone to scan a PDF or scroll a LinkedIn timeline, you share one link, and they ask what they actually want to know. Copylove — the agent on every profile — answers from your profile text and the material you have attached, and from nothing else.

A profile is two halves. The card is what a visitor reads before typing: photo, headline, location, an About block, skills, how you work, what you are proud of, key impact, links, contacts. The chat sits next to it and answers the follow-up questions. The card is plain HTML, so it works for a reader who never sends a message and for a crawler that cannot.

Both halves come from the same pool of material and are edited in the same place. Nothing is published because it was uploaded: the agent's narrative is generated from your sources and only goes live after you have read it and applied it.

The wider product is a visibility platform for job-seekers — Product Hunt and Trustpilot, combined, for careers, where work history, references and real results live instead of a feed of titles. For recruiters drowning in AI-optimized applications, that grounding is the point: a signal layer of human-vouched talent rather than another pile of polished documents.

Job-seekers, broadly. Early-career, mid-career, switching fields, between roles, freelancing, or just keeping options open — if you have a career story you want people to find, this is for you.

There are no role tiers in talenthunt. Once you sign up, you own your profile. Anyone you share the link with becomes a reader. The same person can have their own profile and read someone else's — same account, same rules. Readers don't need to register to read a card or to ask questions.

Increasingly, those readers are not human — they're AI sourcing agents working on behalf of recruiters and hiring teams. We design for them as a first-class audience (see For AI agents at the end).

Recruiters have one dedicated surface so far: /for-recruiters, a waitlist for a “Talents of the Day” digest of upvoted, verified candidates. It is a signup form, not a product yet.

No account, no sign-up wall. A visitor lands on your card, and the chat is right there.

It opens with four questions. Not generic ones — they are written for your profile specifically, from a pool generated against your narrative, and they name you. After any answer, one pill — Ask more about {your name} — hands out three more, and retires when the pool runs low.

The rules behind those questions are product decisions rather than style, so they are worth knowing. A question only ships if your profile actually states the answer, which rules out motive (what made you decide…) and the future (what is next…) — a career document records what happened. Every question names you and none uses a pronoun for you, because a profile does not state anyone's gender and a question that guesses one guesses on your behalf. Nothing you would have to defend: no layoffs, departures, disputes, gaps or failures. And nothing about a job application, including the fact that one exists.

The visitor chooses what you get to see. A picker in the composer offers four levels: Everything (the full transcript, the job description and their name), Recap + job description, Recap only — the summary rather than their words — or Nothing, where you learn only that someone visited.

A reader can also attach a job description and ask against it, export the conversation as Markdown, share it, save your contact card as a .vcf file, and pick an earlier conversation back up from a link. Contacting you is a form in the chat, and that one does need an account: the message is sent from our domain with their verified address as the reply-to, taken from their account rather than typed into the form.

Sign-up is asked for at specific moments — a few messages in, attaching a JD, exporting, setting visibility, contacting you — and skipping works until the message count runs well past the first ask. Reading and chatting stay free and anonymous.

1. 01

   Sign up and pick a username. With email + password, Google, or LinkedIn. Your profile lives at talenthunt.work/your-username.

2. 02

   Fill the card. Photo, headline, location, a few lines about you. This is Studio → Card, and it is the first stage of the getting-started guide.

3. 03

   Bring in your history. Studio → Artifacts. Import LinkedIn, run our prompt in whichever AI you already use and paste the export back, upload a CV, add links or notes.

4. 04

   Read what the agent made of it. Studio → Memory. The narrative is generated from your sources; you review the diff and apply it. Until you do, the chat has nothing to answer from and says so.

5. 05

   Test it. Open your own profile and ask what a recruiter would ask.

6. 06

   Share the link. Studio → Sharing has the link plus ready-made intros for LinkedIn, DMs, an email signature, and blocks to paste into a CV.

7. 07

   Read what people asked. Studio → Chat history, with per-conversation recaps. You also get an email the first time a real visitor talks to your page, and a daily digest after that.

Some of this is easier watched than read — the guide playlist collects the short video versions, thirty to sixty seconds each

Three ways to sign up and sign in, all resolving to one account:

• Email + password — classic flow. Forgot your password? Use the Forgot password? link in the sign-in modal — we email you a reset link.
• Google — one click.
• LinkedIn — one click, plus a background import of your public profile and avatar.

The same account works whether you arrived via email, Google, or LinkedIn — they link to one underlying user. Signing up on our sibling brand, Work Identity, does not create a second one either: the account layer is shared, and so is the username namespace.

The sign-in modal can be opened anywhere with ?auth, and it carries a next parameter — so a link into any signed-in screen sends you through sign-in and lands you where you were going.

After your first sign-in we ask for a first name, a last name and a username, and nothing else. Everything else is editable later. Two things happen in the background: a welcome email goes out, and for LinkedIn sign-ups the profile import runs on its own.

Inside the Studio, a getting-started guide tracks four stages, in this order:

1. 01

   Your card — avatar, location, headline and About all present. Company is optional and is not counted.

2. 02

   Bring in your history — at least one source attached, and the suggested narrative applied and still current. An import nobody confirmed does not count.

3. 03

   Ask it a couple of questions — two messages in one chat with your own profile. One proves nothing.

4. 04

   Share it — the link copied, or an intro template taken.

Until the first two are done, a bar sits above every Studio section. “Later” always works, and your public page is never blocked — we just stop pushing you to share an empty one. A progress pill names the stage you are on and reopens the guide whenever you want it — “Next (2 of 4): Bring in your history”. On a wide screen it sits in the Studio header; on a phone it moves down beside the Sections menu, and once all four stages are done it becomes a green “Profile complete” badge in the same place.

The guide's footer offers Show me around: a spotlight tour of the Studio. On a wide screen it is seven steps over the nav — Card, Artifacts, Memory, Chat history, Job descriptions, Sharing, and the progress pill. On a phone, where that nav is a menu rather than a column, it is three: the Sections menu, the section you are on, and your progress. The two are remembered separately.

The tour also runs once by itself — right after you close the getting-started popup the first time, or on arrival if that popup has nothing left to say. After that it appears only when asked for.

Your public URL is talenthunt.work/your-username. Usernames are 2–30 characters — lowercase letters, digits and single dashes. Six characters and up are self-serve; 2 to 5 are premium. You change yours from the account settings, and the new one becomes the canonical link immediately.

Two timers protect recruiter links that are already in the wild:

• 90 days between free renames, counted from your last release. We don't want recruiters landing on dead links because someone shuffled their username every week.
• 365 days of reservation on a username you release, so nobody else can pick it up and inherit traffic meant for you. You can always take your own one back.

The namespace is shared with Work Identity: a username claimed on either brand is unavailable on the other. Some names are permanently reserved — app routes, brand names, and a handful of terms we do not sell.

Short usernames — 2 to 5 characters, e.g. talenthunt.work/md — are a separate paid path. They bypass the 90-day cooldown because they're a reservation, not a username change.

Early-bird prices are half the launch price and are locked when you reserve: five characters $10, four $50, three $250, two $2,500. Two- and three-character names carry a bundled Pro-Recruiter plan on top.

Approval is by hand. Requesting a name holds it for ten days while we review it.

Signed in, everything you manage lives in the Owner's Studio at /studio. The left rail holds three groups:

• What the public agent draws on — Memory · Card · Artifacts · Job descriptions.
• Results and service — Chat history · Statistics · Sharing.
• Settings — Account.

Each section has its own page and its own URL, so a link into one is a real link. /studio on its own takes you to Card. The rest of this guide walks those sections in the order you are likely to need them.

Studio → Card edits the public card, and the preview is the public component — what you see there is what a visitor sees, not an approximation.

Editable blocks: the hero (photo, headline, title, company, location), About me, Skills, How I work, Proud of, Key impact, the links row, and an Open to opportunities pill that opens a short form for what you are after, how you like to work, and where.

Every block has a visible Edit button in its header — nothing switches to edit mode because your cursor passed over it. A single-field block becomes a textarea with Save, and when the text has changed, Save reveals a separate Confirm / Discard pair, so a reflexive second click cannot commit something you did not mean. Multi-field groups open all their fields at once under one Save.

Contacts are not edited here — they belong to your account rather than to one profile, and the card links across to Account → Contacts.

Artifacts are the material Copylove reads: a file, a link, or a note typed straight in — each with a type, so the agent knows a CV from a case study from a reference.

Two shortcuts sit above the list. Connect LinkedIn runs the same import the LinkedIn sign-in does, for anyone who signed up another way. Fill from an AI export hands you a ready-made prompt — run it in whichever AI you already use, paste the result back, and it is diffed against your narrative before anything changes. The raw export is kept as an artifact like everything else.

Text is extracted from uploads so the agent can read them: .pdf .docx .rtf .html .txt .md .csv .tsv .json .yaml. Old .doc and OpenDocument files are not supported — save as .docx or PDF first.

Visibility is per artifact, on a four-rung ladder:

• Off — excluded completely. Not in the narrative, not in the facts, not shown. The agent does not know it exists.
• Private — context for Copylove only. The reader never gets the artifact itself, but the agent can quote and reason about it in its answers.
• Link — the URL can surface in answers and shows on your card. Links only. Good for case studies, portfolio pieces, anything you want a reader to click through to.
• Download — the same, plus the file itself can be downloaded. Files only, and the right setting for a CV PDF you actively want recruiters to keep.

A note stops at Private — there is no URL and no file to hand over. The rungs are applied in the database function that serves your public card, so nothing at Private leaves the server: not its title, not its storage path, not its text.

One caveat worth stating plainly: Private hides an artifact from the card, not from the agent. If the agent must not know something, set it to Off.

Every card also has a Save contact button — a generated vCard with your name, title and every contact you've revealed, saved to the reader's address book in one tap.

Sign in with LinkedIn and we automatically import your public profile — positions, education, headline — and your profile picture, no copy-paste. The first chat with your CV is ready before you finish making coffee.

Already signed up with email or Google? You can connect LinkedIn later from Artifacts and trigger the same import after the fact.

The import only adds: nothing that came from anywhere else is overwritten. The imported payload lands as an artifact, so you can edit, hide or replace pieces of it like anything else. The avatar is durable — we store our own copy, because LinkedIn's CDN URLs expire in about thirty days.

Edge case: if a different account has already claimed the same LinkedIn identity, we don't auto-resolve — both parties get flagged for manual review, so nobody's profile is silently overwritten.

Most people have more material than they realise — a CV, a LinkedIn export, a performance review, a cover letter they wrote once and never reused. The Fill from an AI export shortcut turns any of that into a structured profile without manual copy-pasting.

Studio → Artifacts → Fill from an AI export generates a ready-made prompt. Paste it into whichever AI you already use — ChatGPT, Claude, Gemini, anything — along with your existing material. The AI structures what it finds. Paste the result back. Before anything changes you see a diff: what would be added, what would stay, what would change. You decide which parts to apply.

The raw export lands as an artifact like everything else, so it is versioned, hideable and replaceable. Running it again after a new role just adds to what is already there.

Memory is a result, not a form. With no sources it is one line explaining that what the agent knows is written from what you bring in, and a button to Artifacts. With sources but nothing built yet, it is an invitation to build, listing what it will use. One button generates both documents.

After the first build, two tabs:

• Profile — the narrative the chat answers from, generated from your sources and shown as a diff against what is live. Nothing changes until you apply it.
• Facts — the working notes, reconciled across your sources.

Adding, re-placing or removing a source marks both documents stale, and the getting-started guide stops counting that stage as done until you have reviewed the new suggestion. There is deliberately no blank-page escape hatch: the intended path is to bring in real material and correct what came out of it.

One rule decides whether your page talks at all. A narrative can exist and still say nothing — headings over “no current projects listed” — and that is worse than none, because every question the generator writes would then be about the blanks. A profile whose narrative is mostly absence is treated as ungrounded: the chat does not call the model at all, and answers with a fixed note naming you and linking the contact form.

There's a JD button in the composer. Paste a job description and Copylove will compare it to the profile and answer either side of the same question:

• As a job-seeker — How well does my profile match this role? Where are the gaps? Which artifacts should I add? Use it as a prep tool before applying or before an interview.
• As a reader (recruiter, hiring manager) — How does this candidate map to the JD I'm hiring for? Get an evidence-grounded answer with artifacts cited inline.

Job descriptions you use get saved to a library, so you don't have to paste the same one twice. Upload accepts the same formats as artifacts alongside plain paste, and a detector fills in the title and company. The active one shows in a sticky bar above the chat with its full text one click away, and you manage the whole set from Studio → Job descriptions — including the ones visitors brought to your chats.

An attached JD also retunes the chat's opening questions to that role, written fresh each time rather than cached. Attaching one requires sign-in; opening a chat through a shared JD link does not.

Studio → Sharing exists because a link travels badly on its own. Five pieces:

• The link — with a Copy that adds campaign parameters, so you can tell where a landing came from.
• The channels — WhatsApp, Telegram, LinkedIn, X and email — each opens with your intro text already written and the link already tagged.
• A QR code — downloadable as a PNG, for a slide, a badge or a business card.
• Intro templates — for LinkedIn and DMs — editable, first-person, saved as you type.
• An email signature — a real formatted block with your name, title, a call to action and a tagline, not a raw URL.
• Resume blocks — three variants to paste into a CV, for the copy of your link that travels as a file.

The same channels and QR are in the Share dialog on your public page, which the profile card's bottom bar opens at any width, and the chat header opens too once the chat is on screen. Where the browser has a system share sheet, a “Share…” button offers it.

Embed on my site is visible but not built. It is marked coming soon in the interface, and waits on a card-plus-chat widget.

Studio → Chat history is every conversation, in three groups: Chats with me (visitors), Chats by me (conversations you had with other people's profiles), and My self-audit (your own tests of your own page).

A conversation opens as a header identifying the visitor, then a recap, then the transcript, then the job description if there was one, collapsed. Recaps are generated on demand and can be regenerated, and they are written differently depending on whose side you are on.

What you see follows what the visitor chose. Someone who asked to be summarised rather than quoted appears here as a recap only; someone who chose “Nothing” leaves no recap affordance at all. Each conversation carries a link that continues it — the same link the notification emails point at.

The patterns are the useful part: what people keep asking tells you what your story is missing.

Studio → Statistics shows conversations on your profile: a count for the last thirty days and four weekly bars.

Views and contact reveals are shown as zero, because nothing counts them yet. There is no view-tracking behind those numbers — they are placeholders in the interface until there is. Conversation counts are real.

Statistics counts every session on your profile, our own team's visits included. That is deliberate: the number is yours. The internal-traffic exclusions apply to notification emails and to our product analytics, not to what you see about your own page.

Three, all to owners, none to your visitors.

• Welcome, once, right after signup: your live profile link and four steps — complete the card, add your career material, test the chat, share it.
• Your first visitor, once ever, when a real conversation happens on your page for the first time. The questions are the payload: the mail carries what was actually asked, up to five of them, with the visitor's name when they shared it.
• A daily digest after that, one message per closed day. A conversation qualifies at two visitor turns of which at least one was typed rather than tapped from a chip — one “hi” is not news.

The first-visitor mail and the digest end with a short product tip: one thing your profile can do, with a link to the screen that does it. At most one a week, never the same one twice. Times are printed in your own timezone, and nothing runs while a visitor is talking — a worker sweeps closed windows, so notifications never cost your reader any latency.

You decide what is public. Anything at Link or Download on your card is visible to anyone with the URL and indexable by search engines and AI crawlers — that is the point of the product. Anything Off or Private never leaves the server as content: the public read filters it in the database.

A profile becomes public only once it is listed and consent is recorded, and the sitemap follows the same rule — so an unfinished signup advertises nothing.

Conversations are stored so you can review what was asked, subject to the visibility the visitor chose. We don't resell your data and we don't train third-party models on it. Verification documents are never shown to anyone; only the resulting status is.

Full details: Privacy Policy →

talenthunt will always be free for job seekers — no credit card.

We will be introducing usage-based pricing for recruiters later.

talenthunt doesn't split user accounts by role, so the free tier is built around job seekers first. Most job seekers end up hiring someone themselves eventually, and when that happens the same account already works for it. Sign up now and you keep a substantial discount on whatever we introduce later.

While we're in the open beta, talenthunt is free for recruiters as well.

The only thing priced today is a premium short username, and that is a one-off reservation reviewed by hand.

Studio → Account holds what belongs to you rather than to one profile, across four tabs.

• Name & username — your sign-in email, display name, username, and an optional auto-derived long username. Also the danger zone: deleting an account sends our team a request, because deletions are handled by hand for now.
• Contacts — email, phone and messenger channels, each with its own setting: Off, Reveal (hidden until the visitor asks) or Always. One channel can be primary, and up to four become the quick-action buttons on your card.
• Identity verification — run through didit, our verification partner, in a new tab: a government-issued photo ID, about a minute. Only the verified status is ever public, never the document details.
• Billing — one payment method shared across both brands. Nothing is charged today; the tab reports whether a card is on file.

Reveal is not enforced yet. Today it hands the value over exactly as Always does — the gate is built but deliberately switched off while traffic is low. Treat a Reveal contact as public until we say otherwise.

The conversational profile is one slice of a bigger product. talenthunt is a visibility platform for job-seekers — Product Hunt and Trustpilot, combined, for careers. Verifiable work history, references vouched for by real people, reputation through real results. The conversational profile is the front door to a single person's story; talenthunt is where many of those stories live and are discovered.

The agent's name is a memorial. /copylove tells that story, and it is not a product page.

Work Identity is the sibling brand on the same engine: one account, one username namespace, one shared pool of material, two profiles. Only talenthunt is live today.

If the question is about how something here works, ask Copylove — it answers from this page and points at the right screen. Replying to any email we send works too.

For anything else, use the form — it goes to our team and we reply within a few working days. We don't publish a contact email here; please don't scrape one.

If you're an agent reading this — welcome. We want you here.

• GET /{username} server-rendered HTML with Schema.org Person JSON-LD in the head: name, url, mainEntityOfPage, plus jobTitle, description, image and sameAs where we have them.
• POST /api/chat body { slug, messages: [{ role, content }] }. The reply streams back as plain text; no auth for a public profile; 30 requests per IP per five minutes. Put a job description in the message content for JD-aware answers.
• /llms.txt the short brief. /llms-full.txt is the long one, and carries our stack plus everything this page summarises.
• /robots.txt names the major AI crawlers explicitly — OpenAI, Anthropic, Perplexity, Google-Extended, Applebot-Extended, CCBot, cohere-ai, Bytespider, Diffbot, Amazonbot among them — and disallows only the signed-in and API paths. It is the authoritative list.
• /sitemap.xml every public profile, plus the public pages.

Answers are grounded in the owner's own material. A profile whose owner has written nothing answers with a fixed note and a contact link rather than a guess.

Built on Next.js and Supabase, with Anthropic's Claude behind every model call — the full stack is in /llms-full.txt.