I made Peekly[0] because I was tired of feeling FOMO about all the stuff happening in AI, dev tools, indie hacking, etc. I couldn’t keep up with blogs, feeds, and newsletters — it was too much.
Peekly pulls from high-quality sources using LLM + retrieval, then sends you a regular digest with just the most relevant content according to your interests. You can even give it custom prompts to control what it finds and how it summarizes — super useful if you want a particular angle on a topic.
YC folks can use code YC256 for an extra free month (on top of the 14-day trial).
Would love to hear what you think!
Building (https://estan.ai), an AI that reverse-engineers real estate listings by pulling not just from public records, but also social media, neighborhood forums, historical sales, and user reviews. It flags red flags—like crime trends, sketchy development patterns, neighborhood trends, and so on—then presents them in plain, opinionated language.
Goal: replace the vague boilerplate of listings with something closer to a paranoid friend who's done way too much research. Curious if this kind of subjective, context-heavy analysis adds real value—or just noise.
Think this is a great idea, I would have paid for this when we were buying our house. Especially for one feature: let me give you a list of properties (perhaps forwarded from my agent, or e.g. a Zillow search) and get back the "enriched" listings with your added insights. It'd be great if I could then weight different factors (local schools, walkability, etc.) to stack rank them.
With a once-in-a-century shakeup in commission structure for Realtors, you're building this at a very interesting time.
Happy to chat more if you'd like to bounce ideas around.
edit: Also, I think you could easily monetize this by charging agents (or their brokers) for it.
I’ve thought about something similar, after having been burned twice by rosy outlook from real estate agents (USA). The issue is there’s no incentives for a tool like this. A pre-sale inspection is really just to protect “the deal” from being unwound, and to check boxes where there is regulation or if one or the other party could be sued post-sale. Neither the selling or buying agent would want a tool that will prevent a deal, and a homeowner may not see the value given they are already paying a ton of cash to their agents. You would need to monetize via some side channel, or convince the buyer to pay you - the non-competitive MLS status quo of US realty would never permit you to partake in a transaction. Or, sell to those who have been burned (Id be a client if I ever can stomach buying a new home again).
Some ideas for your tool:
- noise - neighbors with farm animals (increasingly popular in suburban areas, and local regs haven’t caught up wrt noise), or neighbors who have been reported or fined for animal neglect or noise violations (for example, pets that live outside where homes are close enough for this to be a meaningful noise problem). Problem neighbors in general, which is subjective but probably easy to find outliers.
- traffic patterns, utilization by Waze drivers, speeders, prevalence of stop signs per intersection, if the street is used by emergency personnel or is close to a fire or police station
- frequency of nearby construction via permit applications
- inspector reviews - US inspectors have little oversight and typically protect themselves from damages contractually, it would be good to know which ones regularly flag things like roof leaks or don’t have exclusive relationships with real estate agents (which queers their incentives wrt buyers - “use my guy, he’ll give you a deal” gets contracts signed quickly)
- prevalence of rentals or Airbnb in the area, renters aren’t bad of course but a buyer may want to know there’s a higher density of residents than appears, and all the knock on effects this could cause like traffic, trash, noise, new faces, non-owner residents engaging in risky behavior etc.
I’d considered a device that could be nailed to a nearby telephone pole, like a raspberry pi with a sensitive mic and some presence sensing, with a few days of battery life. A technology version of “hang out in this neighborhood for a bit and give me the low-down that the sellers may want to hide”.
Isn’t a better solution for the given distance - rail transport? It’s more affordable, ecological and can serve higher number of people from different economical layers. This seems more like an enthusiast toy rather than an alternative for 50-300 miles travel options.
It would be nice if the study checked whether the higher amount of individuals in the cancer group got their tattoos AFTER they got diagnosed. This may be the case if people are more open to take risky actions following such diagnoses. It would explain why the size of the tattoo didn’t correlate with change of getting cancer. Such conclusion will ultimately disprove the causation.
I was tired of the need to scroll through dozens of blogs and RSS feeds to learn about technologies and industry news, so I’ve built a service that helps you learn and stay updated about any topic by sending a single fully personalized weekly email digest, making relevant information come to you, instead of you chasing it (push vs pull):
An interesting observation is that Cascadia Subduction Zone is potentially less devastating than Seattle Fault despite being capable of a 100x bigger magnitude (9+ vs 7.2) thanks to a longer distance from the city
If they were going to the amount of energy to do a 2nd crawl using a different user agent, then why bother advertising the user agent at all and just feed it the Chrome one like every other home-grown spider does
Peekly pulls from high-quality sources using LLM + retrieval, then sends you a regular digest with just the most relevant content according to your interests. You can even give it custom prompts to control what it finds and how it summarizes — super useful if you want a particular angle on a topic.
YC folks can use code YC256 for an extra free month (on top of the 14-day trial). Would love to hear what you think!
[0] https://peekly.ai/