I did some protocol reversing and wrote a small program that pretended to be an airplay speaker to pipe audio to a sonos speaker (archive: https://github.com/stephen/airsonos)
I ended up getting recruiting messages from both the airplay team at apple and some folks from sonos. I didn't end up taking either offer, but it was also an interesting talking point when interviewing for the job I did take.
It was a cool project at a time when a lot of people were saying it was insurmountable to make us AirPlay compatible.
Sorry you didn’t get the job. I hope you didn’t lose much sleep over it. I left in 2020. I wouldn’t say you’re missing much any more.
However, what should be used instead? I would take the boxes from Apple, but, as far as I know, they are also quite limited (only AirPlay).
Another example of the fact that such things should rather be expanded with open standards.
From my point of view, the best strategy for Sonos would be to be as open as possible.
This would make for an interesting blog post.
Could you recommended resources to learn to do the same?
A bit light on the technical details perhaps, but I recall getting stuck on getting the right airplay parameters, learning how byte endianness works... happy to try to answer any other questions as best I can remember.
EDIT: Sorry, I realized that I didn't actually answer the other question. I first got interested in reversing from console hacking, specifically this talk about wii hacking: https://youtu.be/0rjaiNIc4W8 (including marcan of asahi linux fame!). Their group also had more writing at: https://fail0verflow.com/blog/. Also interesting to read about mgba emulator development: https://mgba.io/tag/debugging/, v8 internals: https://mrale.ph, react internals: https://overreacted.io/
Consuming a lot of literature on how different systems work helped me develop intuitions around how you might take something apart. Then it's a matter of trying things and banging your head against the wall a lot, e.g. at some point I was interested in how compilers worked so I tried hacking typescript syntax support into babel (circa 2017 maybe) - I got pretty far! and got a lot better sense of how compilers work.
Please excuse the code quality... I think I was still learning how to write js at the time.
Step 2 — deploy the network appliance in question to your LAN and intercept its packets with wireshark.
Step 3 — begin inference of protocol from observed behavior and test hypothesis by sending hand-crafted payloads to the server in question.
Step 4 — rinse and repeat until assumptions are proven to be correct with a high degree of reliability.
A good way to ensure you’ve captured the major parts of the protocol is to record about 72 hours of traffic and then replay it through a proxy that directs traffic to your newly created service.
If you can interpret the vast majority of the messages without error, you’re getting close to a reliable implementation.
Step 5 — use this strategy to develop a deep understanding of both protocols in question.
Step 6 — write an “adapter” that can translate protocol A to protocol B and vice versa.
Step 7 — implement the adapter towards whatever use case you have in mind.
Curious about common tactics people use for avoiding the ban-hammer from the company at this stage. Surely they can tell the difference between normal operations and this kind of hand crafted probing?
I bought a couple and then found 2 pairs of nice second-hand bookshelf speakers. Presumably everyone is upgrading to smart speakers and dumping their old stuff because it was really easy to find great speakers for a very low price.
I added a discord link and they joined that too and just watch and lurk, its a weird world
Lots of people, including me.
I've used Apple products for years and generally like them a lot as a company, but I have no interest in working for them or any other large US tech company.
The money is good, of course, but the quality of life sacrifices aren't worth it (for me).
That's not contrarianism. That's understanding what matters to me.
100% guarantee I would
What’s wrong with compilers?
And imagine how many middle-managers they have who think a) they're the next Steve Jobs, and b) that what drove his success was being a dick to his workmates.
yeah I mean you
https://bemorewithless.com/the-story-of-the-mexican-fisherma...
On a personal level, this is my life now.
https://news.ycombinator.com/item?id=36767599
Do you really think I would give this up in exchange for more money and more stress?
In case it's not, there's an extension of Cunningham's law. If you're wrong and disagreeable to a sufficiently high degree, people will just ignore or quietly ridicule you rather than try and correct you.
Do you seriously think someone cares if you work at FAANG outside of those student groups on Reddit?
So depending on what job you'll want to do at Apple you might have to relocate.
You're doing RTO, and you're signing NDAs out the wazoo.
Helluva opportunity if you're young, but I'm in agreement with the parent posters that it's not the end-all, be-all. I'm far enough along in my career that I don't need the prestige line on my resume, and 100% remote somewhere I want to live is worth far, far more to me.
I say this as someone that’s never worked for any tech company anybody has heard of, nor any hip SV startup.
It takes a particular sort of person to thrive in Big Tech. That isn’t just code for ‘really really really good’. It sounds like you still believe that it is. Plenty of ‘really really really good’ people wouldn’t do well at Apple, and plenty wouldn’t want to work there. I wouldn’t necessarily call myself ‘really really really good’, but I know that I don’t want to work at Apple. Not because I think I’m not good enough, not because k don’t think that I could keep up, and not because I don’t like their output as a company.
Putting companies in a pedestal the way that you are is ultimately damaging for the industry. It fosters the increasingly cringey “get a job at FAANG!!!” culture. I’m sick of my YouTube recommendations being poisoned with “here’s how you pass a system design interview at Google” BS. I implore you to stop putting companies on a pedestal. You need only look at accountants talking about working at the ‘big 4’ to see how utterly ridiculous it can get.
The same is true of Microsoft (Azure specifically), Google, and Amazon.
Only two "traditional" big tech companies of any note for having sharp people and all-around good vibes are Meta and Netflix. Otherwise, I'd rather go with a unicorn like Snowflake or Databricks, which feels more what software engineering was like in the aughts: exploratory, pioneering, actually building things that people haven't before, rather than gluing stuff together or being drowned in the machinations of some incompetent director.
I wouldn't make it at Apple, because I would get pissed off and quit. There's more to life than money. I don't want to work with people that see "FAANG" (or use "staffed by ex-FAANG" in their recruiting pitch) and think it's a good signal to be presenting.
Putting in a stint at any of these lower end tech companies would cost me intangibles that I'm not willing to give up at this stage of my life. Namely, my sanity, spiritual well-being, and fulfillment with life.
I found Apple recruiters very clear and fair.
Apple's recruiters were flaky on comms, rescheduled me a couple of times, and were disappointing for a company I had higher expectations for.
Sure, your experience is probably true, but it’s anecdotal and the sweeping generalizations you’re making on limited datapoints makes things untrustworthy
Second, you’re assuming money is mutually exclusive from all else at those companies. You admit not being able to get in, which makes this the equivalent of “money can’t buy happiness” from someone who never had money to know first hand that they wouldn’t be happy
Here's what I value: getting shit done, being surrounded by intelligent, hard-working, and perceptive people, not being treated like a peon and gaslit to work to the bone. With that in mind, Meta and Netflix are the only ones that satisfy those conditions. Microsoft is working on some very cool stuff, but those are all in their research orgs, while the rest are effectively the same people that would work at Intel: lifers. Not even coasters, but people who don't care that much at all about tech, and are putting in their hours and moving on with their lives. Nothing wrong with that if that's your value system, but it's not mine. Amazon is a neo-feudalist code mill, no different than Infosys. The people there are better than most devs out in the wild, but that's not a high bar. Google lost its touch when Sundar hopped in around 2015. It's not cool. And Apple I've already lambasted. Netflix's and Meta's engineering cultures are great; I like the people; and I like how none of the people I know there are stressed to the core. Though the problems they're working on aren't interesting to me.
I never admitted to not getting in, that's a conclusion you made yourself. I received offers from all of these companies at various points, but chose not to accept.
They’re unicorns for being extremely stupid. I’ll give you that.
Every company has their warts. Some warts don't matter to me.
> Majority wouldn’t want to deal with bullshit at FAANG
Your comment:
> most of us here probably wouldnt be able to keep up with the amount of work it takes to thrive at apple, sure
You have some weird fetish with FAANG?
Well Amazon is a well known shit show (been there done that)
1. Their software gets used by millions of people 2. They are paid incredible amounts of money 3. They have wonderful working environments and perks 4. Are generally given more autonomy than non big tech companies?
Heck, I never wanted to work at a large company. A chance to work remotely at AWS ProServe more or less fell into my lap and from the minute I got there I knew I didn’t want to be there more than four years.
More realistically from my connections and experience I would have a better chance working in the consulting division at GCP (full time position). I wouldn’t work there either even though it would probably pay close to $100K more than I make now.