Claude helps user find lost phone using Bluetooth signal tracking
A developer's story of using an AI assistant to quickly build a phone-finding tool sparks conversation about AI's role in coding.
Conversation activity · last 7 days peak 1/2h
Summary, timeline and people extracted by Claude from 16 items across 2 sources · 1h ago. Quotes are verbatim.
A user posted on X that after losing their phone at the office with Find My disabled, they asked Claude to help locate it. Claude suggested tracking Bluetooth signal strength and wrote a meter in about a minute, which the user used to find their phone. The post went viral on Hacker News, generating discussion about AI's practical value in solving real problems, concerns about code quality, and debate over whether this represents genuine innovation or reinvention of existing solutions.
- Claude rapidly generated a working Bluetooth signal-tracking tool to help locate a lost phone when standard tools (Find My) were disabled, demonstrating practical AI problem-solving.
- Multiple commenters shared similar experiences using Claude for quick technical solutions, ranging from children's games to debugging and protocol reverse-engineering.
- A significant tension emerged between appreciation for AI's speed and skepticism about code quality, with concern that rapid generation creates unsustainable technical debt.
- Community debate reflected disagreement over novelty—some see this as genuinely useful innovation, while others argue Bluetooth tracking is a solved problem already captured in the LLM's training data.
How it unfolded
-
Several commenters pushed back on the novelty of the solution, noting that many apps already do Bluetooth signal tracking and that the LLM was likely trained on existing implementations. One user called it 'no astounding news here.'
“Many apps already do that, and the LLM is trained with them (and a million others). No astounding news here.”
1970-01-01 · Hacker News ↗ -
A commenter noted the seeming contradiction that suggesting LLM solutions receives negative responses on HN, but posting about successfully using an LLM to solve a problem is accepted, pointing to inconsistency in community sentiment.
“Somehow, on HN, suggesting to use an LLM to solve some problem gives you negative responses. But then … posting to HN that you've used an LLM to solve a problem is OK?”
amelius · Hacker News ↗ -
Commenters expressed worry that while Claude can quickly generate working code, the output often becomes a 'spaghetti mess' that will create maintenance burdens. One user noted that 'a lot of code is being written and a lot of it is going to be a real future burden.'
“A lot of code is being written and a lot of it is going to be a real future burden.”
aidos · Hacker News ↗ -
Multiple Hacker News users began sharing their own examples of Claude helping them solve specific technical problems quickly—from building games to debugging GIMP crashes to reverse-engineering Bluetooth protocols.
“Yesterday my 8yo prompted (using voice recognition) Claude to make a game where it would play her a song (say Twinkle) and she had to play it back and get scored…The whole thing took about 15 minutes.”
aidos · Hacker News ↗ -
The X post was shared to Hacker News, where it gained significant attention with 293 points and 215 comments, bringing the conversation to a developer audience.
-
User lost their phone at the office with Find My disabled by MDM. After running out of ideas, they asked Claude how to find it. Claude suggested tracking Bluetooth signal strength and generated a meter in about a minute, which the user used to locate their phone.
“Lost my phone at the office and spent 30 minutes turning the place over. Find My was disabled by MDM. Out of ideas, I asked Claude how I could find it. It suggested tracking the Bluetooth signal strength, then wrote me a meter in about a…”
@un1c0rnioz · X ↗
What people are saying verbatim
“Lost my phone at the office and spent 30 minutes turning the place over. Find My was disabled by MDM. Out of ideas, I asked Claude how I could find it. It suggested tracking the Bluetooth signal strength, then wrote me a meter in about a minute.”
@un1c0rnioz, Original poster · X ↗ · Aug 4, 1:03 PM
“Yesterday my 8yo prompted (using voice recognition) Claude to make a game where it would play her a song (say Twinkle) and she had to play it back and get scored…The whole thing took about 15 minutes.”
aidos, HN commenter · Hacker News ↗ · Aug 7, 6:00 PM
“A lot of code is being written and a lot of it is going to be a real future burden.”
aidos, HN commenter · Hacker News ↗ · Aug 7, 6:00 PM
“I used bluez and bash to lock xscreensaver using some very minimal bash…I think it's really impressive what these agents can do, but you should also consider whether you're asking it to burn tokens reinventing the wheel for you.”
unqueued, HN commenter · Hacker News ↗ · Aug 7, 7:33 PM
“Somehow, on HN, suggesting to use an LLM to solve some problem gives you negative responses. But then … posting to HN that you've used an LLM to solve a problem is OK?”
amelius, HN commenter · Hacker News ↗ · Aug 8, 2:01 PM
“Many apps already do that, and the LLM is trained with them (and a million others). No astounding news here.”
1970-01-01, HN commenter · Hacker News ↗ · Aug 8, 11:51 PM
“I remember wanting something like this, thinking 'this is so dumb, the sensor is here, I just need the code to extract the signal'. Well now you can just prompt for it.”
agenticworldcup, HN commenter, career wireless comms engineer · Hacker News ↗ · Aug 10, 7:01 PM
The conversation positions from the crowd, verbatim
The conversation splits between genuine appreciation for Claude's practical utility in solving real problems quickly, and skepticism about whether this represents meaningful innovation or simply reinvention of existing solutions. Underneath both camps runs a deeper unease about code quality and sustainability when solutions are generated rather than carefully designed.
The dispute Whether the story represents genuine innovation (solving a problem developers couldn't otherwise access) or mere reinvention of existing solutions that the LLM learned from training data.
Claude enables rapid, practical solutions to real problems that developers couldn't easily solve before.
-
“I used bluez and bash to lock xscreensaver using some very minimal bash…I think it's really impressive what these agents can do”
unqueued · Hacker News ↗ -
“I remember wanting something like this, thinking 'this is so dumb, the sensor is here, I just need the code to extract the signal'. Well now you can just prompt for it.”
agenticworldcup · Hacker News ↗ -
“Yesterday my 8yo prompted (using voice recognition) Claude to make a game…The whole thing took about 15 minutes.”
aidos · Hacker News ↗
Claude-generated code creates maintenance problems and represents reinvention of solutions that already exist.
-
“A lot of code is being written and a lot of it is going to be a real future burden.”
aidos · Hacker News ↗ -
“you should also consider whether you're asking it to burn tokens reinventing the wheel for you, or making a pretty wrapper around a wrapper.”
unqueued · Hacker News ↗ -
“Many apps already do that, and the LLM is trained with them (and a million others). No astounding news here.”
1970-01-01 · Hacker News ↗
There's a contradiction in Hacker News culture: suggesting LLMs gets downvoted, but posting success stories gets upvoted.
-
“Somehow, on HN, suggesting to use an LLM to solve some problem gives you negative responses. But then … posting to HN that you've used an LLM to solve a problem is OK?”
amelius · Hacker News ↗
Voices from the web unedited
-
Lost my phone at the office and spent 30 minutes turning the place over. Find My was disabled by MDM. Out of ideas, I asked Claude how I could find it. It suggested tracking the Bluetooth signal strength, then wrote me a meter in about a minute. I walked around watching the
-
I wonder. I remember before emissions, what's underneath a car hood was relatively organized and simple. Then with emissions it became a maze of vacuum hoses and so much other nonsense.then ... in some places (maybe cars that people care about working on) it became cleaner again. In the other places, they added a second hood to hide the mess.There…
-
Yesterday my 8yo prompted (using voice recognition) Claude to make a game where it would play her a song (say Twinkle) and she had to play it back and get scored. The UI was a nice piano with visual feedback. The laptop key served as the keys until I prompted for midi support so she could use the electric piano. The whole thing took about 15…
-
I used bluez and bash to lock xscreensaver using some very minimal bash. It wasn't my idea I believe people on the Gentoo forums were doing it.But you can just loop over something like `hcitool rssi "$MAC"` and project it somewhere, there's a variety of ways.I like using dunstify with the -p option to persist on screen.I think it's really…
-
I have a bad Ethernet cables through unused chimney from my apartament to the basement. The workers I asked to pull it just did it brute force and 2 of pairs are broken. When I was chatting with Claude to see if I can use it somehow to send Ethernet packets one of the Claude responses was to use broken cables as a pilot to pull fiber channel!
-
> A lot of code is being written and a lot of it is going to be a real future burden.This assumes that the models of the future won't find it easier to just throw the code away and rebuild itThis also assumes that the same application build by humans wouldn't become a "spaghetti mess".
-
So GIMP kept crashing on me when importing 48 bit TIFFs. I threw a trace at Codex (without so much as a local GIMP repo) from which it figured out that the culprit was thumb preview handling code. It suggested turning off two non-obvious to me options that would circumvent that codepath it indeed worked.
-
Codex helped me reverse engineer the BTLE protocol for a decibel meter. Now I can visualize the volume on the dance floor from the DJ booth.
-
Started my career in wireless comms back in the day.And I remember wanting something like this, thinking "this is so dumb, the sensor is here, I just need the code to extract the signal".Well now you can just prompt for it.
-
Somehow, on HN, suggesting to use an LLM to solve some problem gives you negative responses.But then ... posting to HN that you've used an LLM to solve a problem is OK?