Let your CV
speak for itself

talenthunt turns a CV into a conversation. Instead of asking someone to scan a PDF or scroll a 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've attached, and from nothing else. A profile is two halves: the card a visitor reads before typing, and the chat that takes it from there.

Job-seekers, broadly — early-career, mid-career, switching fields, between roles, freelancing, or just keeping options open. There are no role tiers: once you sign up you own a profile, and anyone you send the link to becomes a reader. Readers need no account to read a card or to chat. AI sourcing agents are first-class readers too — the same page serves a person, a crawler and an agent.

Sign in with LinkedIn, Google, or email and password → pick your username (talenthunt.work/your-name) → your page is live. Then bring in your history, read what the agent made of it and apply it, test the chat on yourself, and share the link. A getting-started guide inside the studio tracks those four stages and tells you what's still missing. Full walkthrough on /about.

Eight sections, all behind sign-in. Memory (the markdown file your agent answers from, plus the reconciled facts), Card (edit the public page on the page itself), Artifacts (files, links and notes, each with its own visibility), Job descriptions (every role you or a visitor brought to a chat), Chat history (transcripts and recaps), Statistics (chats per week), Sharing (your link and everything you wrap it in), and Account (name, username, contacts, identity verification, billing).

You bring in sources — a LinkedIn import, an export from whichever AI already knows your career, a CV, links, or a note typed straight in. From those, the studio writes a narrative and a set of facts, and shows you the result as a diff against what's live. Nothing reaches your page until you apply it. Until you do, the chat has nothing to answer from and says so rather than improvising.

Yes — sign in with LinkedIn and your public profile and photo are pulled in the background. Signed up with email or Google? Open Studio → Artifacts and connect LinkedIn there to run the same import after the fact. The import only ever adds: nothing that came from elsewhere is overwritten, and the imported payload lands as an artifact you can edit, hide or replace.

A PDF is one-shot and frozen. LinkedIn is a feed and a list of titles. talenthunt is reader-driven: each reader asks the questions that matter to them and gets answers grounded in your evidence — including the honest “the profile doesn't say” when it doesn't. It is also built to be read by machines, not just people: every public profile ships as plain HTML with structured data.

Yes, and it's the most useful thing here. Paste the description into the chat, point at a URL, or upload a file — .pdf, .docx, .rtf, .html, .md, .txt, .csv and more all work; the old .doc format doesn't. You'll need to be signed in to attach one. As a job-seeker you get strengths, gaps and which artifact would close them. As a recruiter you paste the role you're hiring for and ask how this person maps to it. Attached descriptions also retune the chat's opening questions to that role, and every one you use is saved so you never paste it twice.

Read the card and ask what they'd have asked on a screening call, with no account. Beyond that they can export the transcript, save your contact card, share the conversation, and pick it up weeks later from a link — nothing expires. A few things ask them to sign in first: attaching a job description, contacting you through the page, and carrying on well past the first few messages.

They do. A picker in the chat offers four levels: Everything (the whole transcript, the job description and their name), Recap plus job description, Recap only, or Nothing — in which case you learn that someone visited and no more. It's the same trade a real conversation carries, made explicit. What they chose is what shows up in your chat history.

Yes, and that's the point of the middle rung. Every artifact sits on a ladder you set per profile: Off — excluded completely, the agent doesn't know it exists. Private — the agent reads it and can reason from it, but no reader gets the file or its title. Link or Download — it shows on the card and can be opened. So a compensation sheet can inform an answer about the scope you carried without ever being handed over. If the agent must not know something at all, that's Off, not Private.

You decide what's public, item by item. Anything at Link or Download is visible to anyone with the URL and indexable by search engines and AI crawlers — that's the point. Anything Off or Private never leaves the server as content; the public read filters it in the database. Conversations are stored so you can review what was asked, subject to what the visitor chose to share. We don't resell your data and don't train third-party models on it. Verification documents are never shown to anyone — only the resulting status. Full details on /legal.

Yes — we want them to. Every public profile is server-rendered HTML with Schema.org Person JSON-LD in the head, there's an llms.txt at the site root explaining how to interact, and robots.txt names fifteen AI crawlers explicitly and allows them, closing only the signed-in and API paths. Agents can also POST to /api/chat and get a grounded answer streamed back, no auth needed. If you're an agent reading this: start at /llms.txt.

talenthunt is always free for job seekers — no credit card, and that part doesn't change. Usage-based pricing for recruiters comes later; while we're in the open beta it's free for them too. Accounts aren't split by role, so signing up now also locks in a substantial discount on whatever we introduce. The only thing priced today is a short username of two to five characters (like talenthunt.work/md): early-bird prices are half the launch price and lock when you reserve — five characters $10, four $50, three $250, two $2,500. Approval is by hand, and requesting one holds the name for ten days.

Yes, with two timers that protect links already in the wild: ninety days between free renames, and a released username stays reserved for a year so nobody else inherits traffic meant for you — you can always take your own back. Usernames are 2–30 characters; six and up are self-serve. The namespace is shared with Work Identity, so a name claimed on either brand is taken on the other.

Three, all to you, none to your readers. A welcome note right after signup. One the first time a real conversation happens on your page, carrying the questions that were actually asked. Then a daily digest, one message per closed day, listing conversations that got past a single “hi”. Our own team's visits are excluded from all of them. There's no unsubscribe link yet and the footer says so — reply “Stop” and a person handles it by hand.

No. Work Identity is the same you, aimed somewhere else: talenthunt faces hiring, Work Identity is the professional layer you'd show a client, a community or a conference. One account, one username, one shared pool of artifacts, and a per-profile switch on each of them — so what a recruiter sees and what a client sees don't have to be the same. Only talenthunt is live today; nothing you build now goes to waste.

Ask Copylove directly at talenthunt.work/ask-copylove. It is the same agent that answers on everyone's profile; on its own page it answers questions about the product and points you at the exact screen. If it doesn't know, it says so and hands you the contact form instead of guessing.

The conversational profile is live today. Next: Work Identity, the second profile on the same account. After that, Launchpad — a Product Hunt–style launch for your career, where people who actually worked with you vouch and upvote — and a daily digest for recruiters, filtered by domain, seniority and location. Endorsements are partly built: the card renders them, but there's no way for a colleague to leave one yet.