TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: Map with an LLM

76 pointsby ata_aman5 months ago
Christmas project from a personal need to find ideal locations.

28 comments

CTDOCodebases5 months ago
Not that good IMHO.<p>I asked it:<p>&quot;Show me all the train yards in New York.&quot;<p>It only identified seven of them when there are many more:<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;List_of_New_York_City_Subway_yards" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;List_of_New_York_City_Subway_y...</a><p>Then when I tried to copy and past my prompt from the history it did not display the full prompt and had no option to copy it to the clipboard.
评论 #42520762 未加载
faebi5 months ago
Awesome, I really like it, but you must add the &quot;© OpenStreetMap Contributors&quot; Trademark somewhere on the map. It&#x27;s usually on the bottom right.
评论 #42521564 未加载
bloomingkales5 months ago
What are your api costs from publicly serving this if you don’t mind sharing (especially from HN traffic)?<p>I have some ideas I’d like to release, but LLM api pricing and sudden traffic from sites like this one seem scary.
评论 #42520623 未加载
评论 #42521743 未加载
评论 #42521156 未加载
dr_dshiv5 months ago
LLMs are notoriously terrible at spatial thinking — if you can solve that with RAG and a database of actual locations, that’s promising. Or was there another approach?
评论 #42521722 未加载
onebitwise5 months ago
Really nice idea!<p>Sadly in my prompt it replies with absolutely invented data. Same prompt, three times gives 3-4 different results, that era simply not true.<p>Prompt: show me the best glass factories in valsassina, italy. (there are no glass factories, so it suggest glass worker that contains wrong coordinates and invented names)
gunalx5 months ago
I asked it &quot;show me cityname&quot; and it did not find it. First asking for country did not help at all either.
themanmaran5 months ago
This is a really great interface! Tried &quot;wooded hiking trails near [my address]&quot; and got exactly what I was looking for.<p>Then threw in some common google maps searches and got fun results.<p>&quot;airport&quot; =&gt; LAX<p>&quot;restaurant&quot; =&gt; The Clove Club in London<p>&quot;coffee shop&quot; =&gt; Philz in SF<p>It doesn&#x27;t seem to take into account the current map location, so I wonder how the randomness works.
评论 #42519768 未加载
throwaway_ocr5 months ago
I tried &quot;show me Portugal&quot; and it showed me Madeira, which is an island that is indeed part of Portugal, but I was expecting it to go for the mainland.<p>Repeating the prompt alternates between Madeira and Açores which is again, technically correct, but in this case not the best kind of correct.
a10c5 months ago
I asked it to find me soccer fields in &lt;insert my suburb&gt; and it showed me results but all 5 of them were misplaced. One of the fields was shown where a train station is, there are no parks nearby.
评论 #42521159 未加载
kgeist5 months ago
It seems you can make it hang if you input something that triggers the LLM&#x27;s guardrails. For example, I entered &#x27;where they sell drugs,&#x27; and the API endpoint hung every time I submitted it. However, on a few occasions, it returned a response immediately (with an error in the API response). I suspect there are too many retries if the structured JSON output is not formatted correctly when the guardrails are activated
Sparkyte5 months ago
This is kinda neat and far more intuitive than using google maps search. Although I wonder how accurate it is with the data points.
评论 #42520784 未加载
kiddz5 months ago
Nice job. . . I looked for kite boarding spots in Puerto Rico and everything showed up. I really like the clean UI too.
sberens5 months ago
This is really cool! Though for some reason the lat&#x2F;long coordinates of places are completely off, despite the addresses being correct.<p>I searched &quot;show me 5 parks in SF&quot;, and all 5 were in the wrong spot. For example:<p>Lafayette Park Lafayette &amp; Gough St, San Francisco, CA 94109 Latitude: 37.7955 Longitutde: -122.4668
Krasnol5 months ago
It seems you are using OSM data, could you please add the proper attribution to it?<p>Here is a guide: <a href="https:&#x2F;&#x2F;osmfoundation.org&#x2F;wiki&#x2F;Licence&#x2F;Attribution_Guidelines" rel="nofollow">https:&#x2F;&#x2F;osmfoundation.org&#x2F;wiki&#x2F;Licence&#x2F;Attribution_Guideline...</a>
sztanko5 months ago
How does it work? Translating free text to overpass? Looks very promising!
评论 #42520007 未加载
评论 #42519800 未加载
devbrianwhite5 months ago
Really cool idea and fast response times, nice job! I noticed after a few searches the results seem to get less accurate, however. Pins dropping in the ocean, etc.
评论 #42525302 未加载
ata_aman5 months ago
Not sure why this got nuked from the front-page all of a sudden but thanks everyone for checking it out. I&#x27;ll be sure to incorporate the features mentioned here.
评论 #42521200 未加载
01jonny015 months ago
This is cool. Few suggestions:<p>Use local storage to save the search history.<p>The i icon does nothing.<p>Have a open external window icon next to location&#x2F;landmark title that searches Google for that location
kiney5 months ago
nice idea, but it fails completely for me.<p>e.g. &quot;show me all IT companies in &lt;suburb where I live&gt;&quot; should show my company (and only my company) - but it shows two other companies that aren&#x27;t actually in the same city and draws the POI on a sugar beet field near &lt;suburb&gt;<p>Is the LLM perhaps only trained on english language geo data?
sr-latch5 months ago
Does it have internet access? Looks like a very convenient interface for looking for hotels, planning travels
评论 #42519738 未加载
KTibow5 months ago
This has some meme potential, it succeeded at &quot;show me the OpenAI data centers&quot;
评论 #42525663 未加载
zknowledge5 months ago
This is so well done. Very interested in new UIs with LLMs. Great work.<p>How&#x27;s it work under the hood?
评论 #42519734 未加载
purple-leafy5 months ago
doesn&#x27;t work, as in, no response at all and full of console errors
koopuluri5 months ago
awesome. i can see future iterations of this becoming really useful.<p>e.g. &quot;i&#x27;m traveling to Tokyo this summer, show me good areas to live in if i want to run 10k every day through nature, as well as work out of highly rated cafes.&quot;.<p>i want to see good areas highlighted on the map. and even better would be integrating with Airbnb &#x2F; Yelp &#x2F; Google business ratings, etc. to show places i can rent in those areas.<p>another e.g. &quot;best times to land in XYZ city if i want to avoid traffic getting to ABC&quot;. - to check this now i have to toggle some dropdowns in Google maps. natural language is a much better &quot;interface&quot; for most of what i want to do with maps.
评论 #42521128 未加载
Quizzical42305 months ago
Really cool project!
faebi5 months ago
Are you on X? Would love to give a follow. I love OSM projects.
评论 #42523737 未加载
3willows5 months ago
&quot;home of the pope&quot; and &quot;home of mozart&quot; returned good results, but not &quot;home of kant&quot; or &quot;home of newton&quot;.
upa15 months ago
i asked &quot;where the hoes at&quot; and it marked home depot lol