Unusual that an interview captures such a nuanced take, but I appreciate both Ward's pragmatism and his sadness around the death of human-created art. Grateful for the submission.
Deeply curious to know if this is an outlier opinion, a mainstream but pessimistic one, or the general consensus. My LinkedIn feed and personal network certainly suggests that it's an outlier, but I wonder if the people around me are overly optimistic or out of synch with what the HN community is experiencing more broadly.
My impression has been that in corporate settings (and I would include LinkedIn in that) AI optimism is basically used as virtue signaling, making it very hard to distinguish people who are actually excited about the tech from people wanting to be accepted.
My personal experience has been that AI has trouble keeping the scope of the change small and targeted. I have only been using Gemini 2.5 pro though, as we don’t have access to other models at my work. My friend tells me he uses Claud for coding and Gemini for documentation.
I reckon this opinion is more prevalent than the hyped blog posts and news stories suggest; I've been asking this exact question of colleagues and most share the sentiment, myself included, albeit not as pessimistic.
Most people I've seen espousing LLMs and agentic workflows as a silver bullet have limited experience with the frameworks and languages they use with these workflows.
My view currently is one of cautious optimism; that LLM workflows will get to a more stable point whereby they ARE close to what the hype suggests. For now, that quote that "LLMs raise the floor, not the ceiling" I think is very apt.
I think it’s pretty common among people whose job it is to provide working, production software.
If you go by MBA types on LinkedIn that aren’t really developers or haven’t been in a long time, now they can vibe out some react components or a python script so it’s a revolution.
Hi, my job is building working production software (these days heavily LLM assisted). The author of the article doesn't know what they're talking about.
I tend to strongly agree with the "unpopular opinion" about the IDEs mentioned versus CLI (specifically, aider.chat and Claude Code).
Assuming (this is key) you have mastery of the language and framework you're using, working with the CLI tool in 25 year old XP practices is an incredible accelerant.
Caveats:
- You absolutely must bring taste and critical thinking, as the LLM has neither.
- You absolutely must bring systems thinking, as it cannot keep deep weirdness "in mind". By this I mean the second and third order things that "gotcha" about how things ought to work but don't.
- Finally, you should package up everything new about your language or frameworks since a few months or year before the knowledge cutoff date, and include a condensed synthesis in your context (e.g., Swift 6 and 6.1 versus the 5.10 and 2024's WWDC announcements that are all GPT-5 knows).
For this last one I find it useful to (a) use OpenAI's "Deep Research" to first whitepaper the gaps, then another pass to turn that into a Markdown context prompt, and finally bring that over to your LLM tooling to include as needed when doing a spec or in architect mode. Similarly, (b) use repomap tools on dependencies if creating new code that leverages those dependencies, and have that in context for that work.
I'm confused why these two obvious steps aren't built into leading agentic tools, but maybe handling the LLM as a naive and outdated "Rain Man" type doesn't figure into mental models at most KoolAid-drinking "AI" startups, or maybe vibecoders don't care, so it's just not a priority.
Either way, context based development beats Leroy Jenkins.
> use repomap tools on dependencies if creating new code that leverages those dependencies, and have that in context for that work.
It seems to me that currently there are 2 schools of thought:
1. Use repomap and/or LSP to help the models navigate the code base
2. Let the models figure things out with grep
Personally, I am 100% a grep guy, and my editor doesn't even have LSP enabled. So, it is very interesting to see how many of these agentic tools do exactly the same thing.
And Claude Code /init is a great feature that basically writes down the current mental model after the initial round of grep.
Linkedin posts seems like an awful source. The people I see posting for themselves there are either pre-successful or just very fond of personal branding
Speaking to actual humans IRL (as in, non-management colleagues and friends in the field), people are pretty lukewarm on AI, with a decent chunk of them who find AI tooling makes them less productive. I know a handful of people who are generally very bullish on AI, but even they are nowhere near the breathless praise and hype you read about here and on LinkedIn, they're much more measured about it and approach it with what I would classify as common sense. Of course this is entirely anecdotal, and probably depends where you are and what kind of business you're in, though I will say I'm in a field where AI even makes some amount of sense (customer support software), and even then I'm definitely noticing a trend of disillusionment.
On the management side, however, we have all sorts of AI mandates, workshops, social media posts hyping our AI stuff, our whole "product vision" is some AI-hallucinated nightmare that nobody understands, you'd genuinely think we've been doing nothing but AI for the last decade the way we're contorting ourselves to shove "AI" into every single corner of the product. Every day I see our CxOs posting on LinkedIn about the random topic-of-the-hour regarding AI. When GPT-5 launched, it was like clockwork, "How We're Using GPT-5 At $COMPANY To Solve Problems We've Never Solved Before!" mere minutes after it was released (we did not have early access to it lol). Hilarious in retrospect, considering what a joke the launch was like with the hallucinated graphs and hilarious errors like in the Bernoulli's Principle slide.
Despite all the mandates and mandatory shoves coming from management, I've noticed the teams I'm close with (my team included) are starting to push back themselves a bit. They're getting rid of the spam generating PR bots that have never, not once, provided a useful PR comment. People are asking for the various subscriptions they were granted be revoked because they're not using them and it's a waste of money. Our own customers #1 piece of feedback is to focus less on stupid AI shit nobody ever asked for, and to instead improve the core product (duh). I'm even seeing our CTO who was fanboy number 1 start dialing it back a bit and relenting.
It's good to keep in mind that HN is primarily an advertisement platform for YC and their startups. If you check YC's recent batches, you would think that the 1 and only technology that exists in the world is AI, every single one of them mentions AI in one way or another. The majority of them are the lowest effort shit imaginable that just wraps some AI APIs and is calling it a product. There is a LOT of money riding on this hype wave, so there's also a lot of people with vested interests in making it seem like these systems work flawlessly. The less said about LinkedIn the better, that site is the epitome of the dead internet theory.
Just wanted to say thanks for the submission. I never would have found this story otherwise, and it's both a powerful read and especially relevant to a close friend—she was one of the airlifted/trafficked babies and has never been able to find her birth parents.
While many of the images were low resolution or just strange icon graphics, and the text felt like AI-slop, the core concept is great. Hope you keep making progress on this.
Hey thanks! The image issue is something i'm struggling with, I source them from Wikitravel and they are often not high-resolution.
Scraping from google images search introduces two (potential) issues:
1. Getting the correct location, (there is for example a Barcelona in Venezuela) but searching that will always just return the one in Spain, the feed should show obscure locations so this is a problem
2. There seems to be some more legal implications from scraping/using images from Google Image search
I'll work on the text feeling like AI-slop, I welcome all the feedback!
Incredible work, Brian! I'm in awe of what you've done in such a short amount of time. I started an indie game company (still working on our first title) and had a few related questions if that's OK?
1) If you didn't need a salary or marketing help, would you still have signed a publishing deal? My sense is most of the publisher value lies in getting paid before the game launches, and with marketing around launch, but curious if those are wrong assumptions?
2) Early Access vs. Straight Launch - any insights about why you chose to do a full launch vs. an early access beforehand? Was it something you and the publisher discussed in detail?
3) Outside of Steam's ecosystem, how much marketing, promotion, social media, YouTube, Discord, etc. stuff did you (or the publisher) do? Do you think that pre-launch work had a sizable impact on the launch and post-launch success? Or would you say it's mostly the Steam algo making the game more visible to the right buyers as the positive reviews rolled in?
Truly kind of you to share so openly here. Already sent some of your other replies to our game team
1) It obviously varies w/ each dev's situation, but I think your sense of the value prop is a fair default one w/o context. In my case, there were four reasons: 1) having a partner to help with (to me) "the unknown unknowns" (i'm only a dev and a novice designer, literally anything else would be my first time doing it, so i figured having an experienced partner there would be wise) 2) the advance was nice just from a "bird in the hand" mindset 3) they helped me connect with a really good artist (critical to the project's success imho) 4) having a big name behind you can't hurt, and bigger publishers do have more relationships with the ecosystem (streamers, platforms, etc) to help your game succeed.
2) EA is appropriate for certain kinds of games. But I knew exactly what Ballionaire was going to be, and felt it could be achieved in a year. I knew that this mechanical space was going to be rapidly saturated, once all the games inspired by LBAL (including Balatro!) started appearing. So I was determined to get the game out before that happened, which set a certain scope and pace, obviating EA.
3) The trailers were a really good promotional tool, very effective. But most of the paid promotion was sponsored streams, not traditional "marketing" (ads). I think the game has a natural tendency for organic spread, due to its fun/simple premise, watchability/streamability, low price point, and so on. It's just an easy game to see, and say "ooh, I wanna try!" because it's instantly understood how to play, and IMHO is very inviting aesthetically.
My takeaway: Make sure your hook is glowingly radioactively good. Don't overbalance. Leave in some jank. Scope down and finish quickly. And avoid tropes. Stand out. (Of course this only works for a certain kind of game!)
"Make sure your hook is glowingly radioactively good. Don't overbalance. Leave in some jank. Scope down and finish quickly. And avoid tropes. Stand out."
Sounds like outstanding advice. I hope we can follow in your footsteps! (and thanks for the kind and comprehensive answer)
How far were you into building the game before you began to approach publishers? YC speaks frequently about investing in the founders, and not the idea---but then dives right into well you need to have revenue and/or users and it's a bit of a double-standard---so I'm curious what stage you were at when you started talking to publishers and what state your product was in when you finally signed a deal.
I'm curious about this as well so I took a look at the publisher's site and found some info: https://rawfury.com/talk/ under "Game Pitches". Would still like to hear from OP on any specifics though.
RE: "Make sure your hook is glowingly radioactively good. Don't overbalance. Leave in some jank. Scope down and finish quickly. And avoid tropes. Stand out."
For games with a limited run time is it possible to balance this against Steam's 2 hour return policy? Feel like we're now trapped in a position where you can make short little proof of concepts on itch but when it comes to trying to make something professional you have to bloat it in a way to minimise people abusing that.
This is very much personal opinion but if you can blast through all of a games content within 2 hours of purchase it wasn't finished / big enough to justify a financial transaction in the first place.
Two hours is about movie or novella worth of content. Both are priced at about the same rate. Not all games have to have 10s or 100s of hours of content. Today, I expect a big(ger) part of gamers are like me: older people who like games, but can only game a few hours a week. Last non-4X game I fully played was CoD2. That was 2005.
Honestly, that could go both ways. Like I'll happily pay cost-of-small-game to watch a movie that will be over in two hours, and there's a whole category now of single-sitting games like Journey, Minit, A Short Hike, etc, that are easily able to be completed in that timeframe but are obviously worth their modest purchase price.
To some of us adult gamers with actual lives and commitments, having something that can be completed in 1-4 sittings is a huge boon, like okay this is a thing I can do without abandoning my family for the next two weeks.
Were you trying to make short games or offer some kind of abridging service?
Because I've wondered about that too. Like, right now there's a pretty big dichotomy between "purchase and play a game myself" and "watch someone else play it on youtube/twitch". But it would be interesting if there was a market for a kind of interactive guided tour, like for $10 let me play the best 4-6 hours of AC:Odyssey, and that's delivered as a mod that just trims all the fat, levels me up quickly enough to hustle through the main story beats and see the good boss encounters and action set pieces.
(Ubisoft themselves sort of do this with their paid XP boosters, but that doesn't actually cut content, it just lets you skip a few hours of grinding over what is otherwise still a 40hr+ experience for most people)
Perhaps I'm an outlier. I expect to pay for my personal gaming experience. But if there's some necessary part of gameplay I don't like, that's negative experience that makes the game worse for me, reducing the value of the game to me. To skip that gameplay seems like something that shouldn't cost me anything, or even get me a discount because I'm not getting some of the experience I paid for. Like say if a side dish is so bad I send it back at a restaurant. I neither expect to pay for that nor a premium for someone else to eat it for me!
So I'm not willing to pay a premium for such a thing. I don't see why the game with the bad part missing should be worth more than with the bad part present. Rather, the inverse! I'll more likely skip the game entirely and find a different one that doesn't have such mandatory grinding.
As I say maybe that's an outlying opinion since making money from this kind of thing apparently works. It it helps, I'm in the adult with family demographic and my time (rather than game purchase levels of money) is what is at a premium.
I've never bought an XP booster myself, and I feel some of the same conflict. Although I can obviously afford it, I don't think I could stomach doing this for new releases. So at most it would be something where I'd be interested after 2-3 years when all the patches are in, bugs are fixed, and review consensus has settled.
So rather than pay US$70 today for a buggy, grindy new release experience, I pay $20 in two years for the base game + $10 for the "player's digest" mod.
I expect even then it would be a tough sell, particularly having to be on PC— a lot of the market for this kind of thing would more casual console/mobile/streaming players.
I love this idea; I can no longer justify 40-100 hour playthroughs. Lots of people rave about Elden Ring but the time investment is just really offputting. I'd buy a version of it at half the price and one quarter of the playtime.
That said, I'd imagine a fair chunk of the impact of major set pieces and end bosses is that you've spent many hours to get to that point. An abridged version would have to leave bough in to allow emotional investment in the story. This is still possible.
Gamers Digest? (For those of you who remember Reader's Digest).
I was making short games. Specifically I was trying to make games that packed the same density of concentrated entertainment as other media.
The abridgement thing sounds like a smart approach, but ultimately I've gotten more enjoyment from shorter games that don't pack the fat in to begin with.
My PM perspective on this is that games are dominating the media market primarily because they are a good value proposition in our tightening economy. I think a good thing to shoot for is around 2 hours of game play per dollar spent, e.g. make it so that it's an absolute no-brainer for a customer to buy your game vs going to a movie. Replayability is a major aspect of this, look at how many hours people spent on LBAL and Vampire Survivors despite both having relatively simple mechanics.
A big part of the issue is that a lot of people do not consider it abuse. The value of games is so heavily burned into a lot of people's minds as being a dollars to hours ratio and Steam having this blanket policy could be perceived to be endorsing that perspective.
> My sense is most of the publisher value lies in getting paid before the game launches, and with marketing around launch, but curious if those are wrong assumptions?
Don't publishers pay you a cut AFTER sales start coming in?
reply