Hemklok is an AI buyer's agent built for the Swedish housing market. You describe what your life needs — we read the listings, the BRF reports, the commute, the history — and tell you what's actually worth your time.
A conversation, not a filter form. "I have a dog and a remote job, want under 30 minutes to T-Centralen, hate hearing my neighbors." That's enough to start.
Listings, BRF annual reports, slutpriser, commute times, planned construction nearby. The boring homework you should be doing but probably aren't.
A short, honest list. Why each one fits or doesn't. What to ask the broker. What the BRF is hiding in note 14. Where the risks are.
We started Hemklok because buying an apartment in Sweden is harder than it should be — not in the way real estate is hard everywhere, but in a specifically Swedish way. The listings live on one site. The BRF's real financial health lives in a PDF nobody reads. The slutpris tells one story; the actual neighborhood tells another. And somewhere in the middle of all that, a person is supposed to make one of the biggest decisions of their life in three weekend viewings.
So we're building the thinking partner we wished we'd had. A son and a father — different generations, the same frustration. One of us is closer to the technology, the other closer to the years of having watched friends and family stumble through this process. We think there's something between another listings site and an expensive buyer's broker, and we think AI is finally good enough to be that thing.
We're early. The site you're reading is a placeholder for something real. If you're someone who's looking for an apartment in Stockholm right now, or someone who builds in this space, we'd genuinely like to hear from you.
The first version of the agent — a real conversation that figures out what you actually need. "I have a dog" means more than just "pets allowed." Try the beta →
Reading the årsredovisning the way a careful buyer would. Debt per square meter. Maintenance backlog. Note-14 surprises. Renovation plans the seller forgot to mention.
Real travel times to the places you actually go — work, family, the gym — using SL and Trafiklab data, not a Google Maps best-case from the front door.
Honest analysis of what comparable apartments actually sold for, what the neighborhood is doing, and where the value is likely going — without the magical-thinking optimism most price estimates carry.
The agent doesn't go to sleep. New listings get evaluated against your preferences, and you only hear about the ones that genuinely beat what's already on your shortlist.
Drafts the questions, manages the follow-ups, tracks the responses. You approve, we send. Human in the loop, less inbox in your day.
Personalized to the specific apartment — what to look for, what to ask, what a fair counter-offer looks like given the comparable sales and the BRF situation.
The other half of the problem. If we can help buyers make smarter choices, we can help sellers price honestly and present clearly. Coming after the buyer experience is solid.
The typical process means searching Hemnet or Booli, attending weekend viewings, and bidding within a few days. The hard part is not finding listings — it is the careful work between search and decision: reading BRF annual reports for hidden risks, comparing the asking price to recent slutpriser, computing realistic commute times, and tracking new listings continuously. Hemklok handles that layer.
Hemklok is an AI buyer's agent for the Swedish residential property market. You describe what you need in plain language; the agent evaluates listings, reads BRF financial reports, calculates real commute times, and presents a short ranked list with honest reasoning. Built specifically for Sweden, starting in Stockholm.
Hemnet and Booli are listings marketplaces — they show you what is available. Hemklok is a buyer's-agent layer above them: it tells you which listings are right for you, why, and what risks each carries.
When you buy a Swedish bostadsrätt, you also acquire a stake in the building's cooperative. BRF debt under 6,000 SEK per m² is conservative; over 12,000 SEK signals elevated risk. This is filed publicly with Bolagsverket but most buyers don't read it carefully.
Stockholm first, then greater Stockholm (Solna, Nacka, Lidingö, Sundbyberg, Danderyd, Täby), then Gothenburg and Malmö.
A father-and-son team based in Stockholm. The project is developed openly at github.com/llm-brf.