Hacker Newsnew | past | comments | ask | show | jobs | submit | agent281's commentslogin

I understood their comment as going from

$100 / hour * 100 hours

to

$100 / hour * 500 hours

not to

$500 / hour * 100 hours


Yeah the last one. The others would require 5x deal flow which LLMs might not help deliver at all. But the last one should exist for people if 10x is true. Not every client can have already fully price in LLM improvements, people have contracts negotiated pre LLM. I have not heard of this though so I have to remain sceptical


But it specifically mentions having 5x the previous hourly rate.


I appreciate this example. This does seem like a pretty difficult feature to build de novo. Did you already have some machine vision work integrated into your app? How are you handling machine vision? Is it just a call to an LLM API? Or are you doing it with a local model?


There was no machine vision stuff in the app at that point. Claude suggested a couple of different ways of handling this and I went with the easiest way: piggybacking on the Apple Vision Framework (which means that this feature, as currently implemented, will only work on Macs - I'm actually not sure if I will attempt a Windows release of this app, and if I do, it won't be for a while).

Despite this being "easier" than some of the alternatives, it is nonetheless an API I have zero experience with, and the implementation was built with code that I would have no idea how to write, although once written, I can get the gist. Here is the "detectNodWithPitch" function as an example (that's how a "nod" is detected - the pitch of the face is determined, and then the change of pitch is what is considered a nod, of course, this is not entirely straightforward).

```

- (void)detectNodWithPitch:(float)pitch { // Get sensitivity-adjusted threshold // At sensitivity 0: threshold = kMaxThreshold degrees (requires strong nod) // At sensitivity 1: threshold = kMaxThreshold - kThresholdRange degrees (very sensitive) float sens = _cppOwner->getSensitivity(); float threshold = NodDetectionConstants::kMaxThreshold - (sens * NodDetectionConstants::kThresholdRange);

    // Debounce check
    NSTimeInterval now = [NSDate timeIntervalSinceReferenceDate];
    if (now - _lastNodTime < _debounceSeconds)
        return;

    // Initialize baseline if needed
    if (!_hasBaseline)
    {
        _baselinePitch = pitch;
        _hasBaseline = YES;
        return;
    }

    // Calculate delta: positive when head tilts down from baseline
    // (pitch increases when head tilts down, so delta = pitch - baseline)
    float delta = pitch - _baselinePitch;

    // Update nod progress for UI meter
    // Normalize against a fixed max (20 degrees) so the bar shows absolute head movement
    // This allows the threshold line to move with sensitivity
    constexpr float kMaxDisplayDelta = 20.0f;
    float progress = (delta > 0.0f) ? std::min(delta / kMaxDisplayDelta, 1.0f) : 0.0f;
    _cppOwner->setNodProgress(progress);

    if (!_nodStarted)
    {
        _cppOwner->setNodInProgress(false);

        // Check if nod is starting (head tilting down past nod start threshold)
        if (delta > threshold * NodDetectionConstants::kNodStartFactor)
        {
            _nodStarted = YES;
            _maxPitchDelta = delta;
            _cppOwner->setNodInProgress(true);
            DBG("HeadNodDetector: Nod started, delta=" << delta);
        }
        else
        {
            // Adapt baseline slowly when not nodding
            _baselinePitch = _baselinePitch * (1.0f - _baselineAdaptRate) + pitch * _baselineAdaptRate;
        }
    }
    else
    {
        // Track maximum delta during nod
        _maxPitchDelta = std::max(_maxPitchDelta, delta);

        // Check if head has returned (delta decreased below return threshold)
        if (delta < threshold * _returnFactor)
        {
            // Nod complete - check if it was strong enough
            if (_maxPitchDelta > threshold)
            {
                DBG("HeadNodDetector: Nod detected! maxDelta=" << _maxPitchDelta << " threshold=" << threshold);
                _lastNodTime = now;
                _cppOwner->handleNodDetected();
            }
            else
            {
                DBG("HeadNodDetector: Nod too weak, maxDelta=" << _maxPitchDelta << " < threshold=" << threshold);
            }

            // Reset nod state
            _nodStarted = NO;
            _maxPitchDelta = 0.0f;
            _baselinePitch = pitch;  // Reset baseline to current position
            _cppOwner->setNodInProgress(false);
            _cppOwner->setNodProgress(0.0f);
        }
    }
}

@end

```


IMO, Haskell is less helpful for an LLM because of its advanced language features. The LLM is reasoning about the language textually. Since Haskell is very tense, the LLM would need a very strong model of how the language works.

I think languages with more minimal features and really good compile time errors would work well with LLMs. In particular, I've heard multiple people say how good LLMs are at generating Go.

Personally, I like languages with type inference so this wouldn't be my preference.


Have you heard of the War of the Currents?

> As the use of AC spread rapidly with other companies deploying their own systems, the Edison Electric Light Company claimed in early 1888 that high voltages used in an alternating current system were hazardous, and that the design was inferior to, and infringed on the patents behind, their direct current system.

> In the spring of 1888, a media furor arose over electrical fatalities caused by pole-mounted high-voltage AC lines, attributed to the greed and callousness of the arc lighting companies that operated them.

https://en.wikipedia.org/wiki/War_of_the_currents


Tesla is barely mentioned in that article which is somewhat surprising


Not surprising at all. He was a minor player in the Current Wars compared to his primary benefactor, George Westinghouse. His image was rehabilitated first by Serbian-Americans and then by webcomic artists and Redditors, who turned him into a secular saint.

Most of what people think they know about Tesla is not actually true if you examine the historical record. But software engineering as a discipline demands business villains and craftsman heroes, and so Edison and Tesla were warped to fit those roles even though in real life there is only evidence of cordial interactions.


Maybe? But it destabilized the Middle East, caused the migrant crisis in Europe, the migrant crisis caused a rise in right wing movements in Europe, it caused the rise of ISIS (lots of Iraqi ex-military), ISIS was involved in the the civil war in Syria, it caused a loss of faith in the American government, created a generation of disillusioned combat vets, so on and so forth.

I really think we're still recovering from the damage caused by Bush administration.


Are we recovering? The knock-on damage you list seems to be accelerating if anything.


Fair enough!


I agree with GP. I would amend their claim with "most problems* could be solved by building high density housing and services in areas with jobs." I.e., build real cities.

Building homes on federal land in the middle of no where will not do anything for people. We just need to allow people to build housing where there is a demand for labor.

Some things I think would be solved include:

- the housing crisis

- mobility => it would be easier for people to move to other parts of the country because they would be less tied to their homes - labor mismatches

- climate change => less reliance on cars

- funding infrastructure => more dense infrastructure means you don't have as much infrastructure to repair and you have more people paying for it

- city government budgets => high density areas are more tax efficient

- home insurance => the homes on the outskirts of cities are most likely to burn down; if housing is cheap the cost to insure it will be cheaper as well

IMO, if housing is 30-60% of peoples budgets and transportation is another 10-20%, if you can bring those costs down you can de-stress a lot of people. That might make politics less intense too.

* "Most problems" is not strictly accurate. But "more problems than you might think are directly related to housing" doesn't really roll off the tongue.


I live in America. I have all my life. I've ridden on nice trains in Europe and Japan. The culture in the last 80-100 years has revolved around cars, but before that relied heavily on trains.

I much prefer riding a spacious train to riding in a cramped airplane. I like to be able to get up and walk around while the train is traveling. I prefer the minimal security processes on a train to an airplane.

Driving 6-10 hours is pretty miserable. I would much rather take a train or fly unless there are specific reasons why I need to drive.

It may be like comparing apples and oranges, but some people prefer apples. The good thing is that you can still eat oranges if you like.


Right but have you flown for business every few days like that? It seems like you’re still thinking about leisure travel between LA and SF.

My whole point was about business flight. Sure for leisure I’ll take a train. But if I’m flying for work I’d rather deal with a consistent (mostly, compared to any rail I’ve taken) experience.


I would not describe LA to SF as a consistent experience. I had an 8:30 PM flight delayed until 1:30 AM. Actually, the only consistency I've had with those flights is that they are consistently delayed.

I've heard that the flight delays cascade throughout the day. So if the first flight is 10 minutes delayed, all SF-LA flights are delayed for the rest of the day. Since SFO has a lot of fog, my understanding is that it's often delayed.

Maybe it's okay coming from Burbank, but I live on the Westside so I have to take LAX.

I would love to have another option. Ideally, a train would decrease the load on the airlines so that they could handle delays better.

To answer your question directly, no I don't have to do work trips every couple of days, but I do have work trips from LA to SF every couple of months.


California can't raise property taxes unless Prop 13 is repealed and that's a very high bar to pass.


A lot of retired people would lose their houses so it would be a political disaster to remove it.


Yeah, I think we should keep that aspect of prop 13. That is what it was originally billed as.

It just doesn't make sense that golf courses pay taxes that are more aligned with the 70's property values.

If there was ever a major change to prop 13 taxes, I think we would need to phase in the new taxes over a five to ten year period. Otherwise, the market would be thrown into chaos.


I mean, majority vote for a certain candidates should do it. If enough people bothered to vote, it could happen.


You would need to overturn the proposition directly. The state Congress can't overturn it so you would need another proposition. That's a "simple" majority vote, but it's the third rail of California politics. We almost got it removed for businesses, but COVID happened and people were not in the mood for raising taxes.


Porque no los dos?


Last time I checked (2020) there were already a few rejected proposals to shorten the syntax for this. It seemed like they were pretty exasperated by them at the time.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: