Hacker News Reader

exhazeon OpenAI O3 breakthrough high score on ARC-AGI-PUB

Hey, that's a great question. I should have been more clear: for deterministic generation that's not done using an LLM. It's done using just regular execution of TypeScript. The code generators that were created using an LLM and that I manually checked for correctness, they're the ones that are generating the other code - most of the code. So that's where the determinism comes in.

exhazeon OpenAI O3 breakthrough high score on ARC-AGI-PUB

You misunderstand the fundamentals. I've built a type-safe code generation pipeline using TypeScript that enforces compile-time and runtime safety. Everything generates from a single source of truth - structured JSON containing the business logic. The output is deterministic, inspectable, and version controlled.

Your concerns about mysterious AI code and system crashes are backwards. This approach eliminates integration bugs and maintenance issues by design. The generated TypeScript is readable, fully typed, and consistently updated across the entire stack when business logic changes.

If you're struggling with AI-generated code maintainability, that's an implementation problem, not a fundamental issue with code generation. Proper type safety and schema validation create more reliable systems, not less. This is automation making developers more productive - just like compilers and IDEs did - not replacing them.

The code works because it's built on sound software engineering principles: type safety, single source of truth, and deterministic generation. That's verifiable fact, not speculation.

exhazeon Launch HN: Human Layer (YC F24) – Human-in-the-Loop API for AI Systems

I am okay with a counterfactual alternate future where some disproportionately powerful entity squeezes Python out of the market: Big TypeScript - funded by a PAC. Offshore accounts. Culprit: random rich Googler who lost an argument to Guido Van Rossum 10 years ago.

exhazeon Launch HN: Human Layer (YC F24) – Human-in-the-Loop API for AI Systems

This really makes you take a step back and just consider the world we're in now: someone critiques a company's approach as unscalable because...

"hiring humans is a nightmare"

Good LLord

exhazeon Launch HN: Human Layer (YC F24) – Human-in-the-Loop API for AI Systems

This is something we frequently saw at Uber. I would say it's the same as there's already an established pattern for this for any sort of destructive action.

Intriguingly, it's rather similar to what we see with LLMs - you want to really activate the person's attention rather than have them go off on autopilot; in this case, probably have them type something quite distinct in order to confirm it, to turn their brain on. Of course, you likely want to figure out some mechanism/heuristics, perhaps by determining the cost of a mistake, and using that to set the proper level of approval scrutiny: light (just click), heavy (have to double confirm via some attention-activating user action).

Finally, a third approach would be to make the action undoable - like in many applications (Uber Eats, Gmail, etc.), you can do something but it defers doing it, giving you a chance to undo it. However, I think that causes people more stress, so it’s rather better to just not do that than to confirm and then have the option to undo. It’s better to be very deliberate about what’s a soft confirm and what’s a hard confirm, optimizing for the human in this case by providing them the right balance of high certainty and low stress.

exhazeon Excalidraw's AI Tool – useful and underrated

I checked out the poster:

1. They don't seem to have any affiliation with the company

2. Excalidraw doesn't seem to have any sort of affiliate program

3. "Why is Excalidraw so good?" - https://news.ycombinator.com/item?id=29109995

I think a lot of people just like it that much

exhazeon AI crawlers need to be more respectful

Having built an AI crawler myself for first party data collection:

1. I intentionally made sure my crawler was slow (I prefer batch processing workflows in general, and this also has the effect of not needing a machine gun crawler rate)

2. For data updates, I made sure to first do a HEAD request and only access the page if it has actually been changed. This is good for me (lower cost), the site owner, and the internet as a whole (minimizes redundant data transfer volume)

Regarding individual site policies, I feel there’s often a “tragedy of the commons” dilemma for any market segment subject to aggregator dominance:

- individual sites often aggressively hide things like pricing information and explicitly disallow crawlers from accessing them

- humans end up having to access them: this results in a given site either not being included at all, or accessed once but never reaccessed, causing aggregator data to go stale

- aggregators often outrank individual sites due to better SEO and likely human preference of aggregators, because it saves them research time

- this results in the original site being put at a competitive disadvantage in SEO, since the their product ends up not being listed, or listed with outdated/incorrect information

- that sequence of events leads to negative business outcomes, especially for smaller businesses who often already have a higher chance of failure

Therefore, I believe it’s important to have some sort of standard policy that is implemented and enforced at various levels: CDNs, ISPs, etc.

The policy should be carefully balanced to consider all these factors as well as having a baked in mechanism for low friction amendment based on future emergent effects.

This would result in a much better internet, one that has the property of GINI regulation, ensuring well-distributed outcomes that are optimized for global socioeconomic prosperity as a whole.

Curious to hear others’ perspectives about this idea and how one would even kick off such an ambitious effort.

exhazeon Ask HN: Is there a list of grey-hair-friendly IT companies?

> I actively hate startups

Have you asked yourself why you hang out on this website given how you feel about startups?

Also, is it really fair to apply such a strong statement to all startups? Look at your screen and the browser you're using, the user interface concepts, the computer itself. Where do you think those came from?

exhazeon TotalRecall: Extracts and displays data from the Windows 11 Recall feature

https://techcrunch.com/2023/04/24/greywing-seagpt/

This helps container ships operate. Errors matter here. They solved them. Please show me where the scam is here?

exhazeon Ask HN: The "Japanese Developer" Scam

If you want to really double check - tell them "it's cool how you guys also take the plastic ring off plastic bottles, not just the cap"

If they say yeah we do that that is 100% confirmation that they do not live in Japan.

Super tech-specific one - ask them you want to make sure they're Japanese so you'll ask them to write something every developer can write in... 5-7 seconds max. When they confirm and are waiting, ask them "at your previous/current company or your friends' companies, how do people name their personal channels? what is the specific kanji ppl put at the end?"

This kanji is super hard to google and even harder to write phonetically, but every Japanese engineer who has ever used Slack knows how because there's almost like a "hack" for writing certain other characters that end up getting converted into that kanji.

This is so fun - 1 more:

Ask them what their favorite hummus flavor is. Basically no one in Japan (even people who have lived outside of Japan) knows what hummus is (I guess Slack even tried to teach them by making it a notification sound lol)

exhazeon Ask HN: The "Japanese Developer" Scam

If a Japanese person who says they live in Japan contacts you and speaks English that isn't super broken, you can safely assume it's a scam. Roughly 2% of Japanese nationals living in Japan can speak English at this level.

exhazeon What we've learned from a year of building with LLMs

I believe we were discussing the former not the latter? I agree that for lots of problem solving tasks it can be hit or miss - in my experience, all the models are quite bad at writing decent frontend code when it comes to the rendered page looking the way you want it to.

What you're describing is more about reasoning abilities - that's not really what the article was about or the problems the techniques are for. The techniques in article are more for stuff like Q&A, classification, summarization, etc.

exhazeon What we've learned from a year of building with LLMs

Please just go and try the lumentis library I mentioned - that is what was used to generate this. It works. For the library docs I showed, I literally just wrote a zsh script to concat all the code together into one file, each one wrapped with XML open/close tags, and fed that in. Just because you weren't able to do it doesn't mean it's a novelty.

exhazeon What we've learned from a year of building with LLMs

https://github.com/hrishioa/lumentis

Since you seem to have not noticed my comment above, here's another example of a project that implements many of these techniques. Me and many others have used this to transcribe hour long videos into a well organized "docs site" that makes the content easy to read.

Example: https://matadoc.vercel.app/

This was completely auto-generated in a few minutes. The author of the library reviewed it and said that it's nearly 100% correct and people in the company where it was built rely on these docs.

Tell me how long it would take you to write these docs. I'm really confused where your dismissive mentality is coming from in the face of what I think is overwhelming evidence to the contrary. I'm happy to provide example after example after example. I'm sorry, but you are utterly, completely wrong in your conclusions.

exhazeon What we've learned from a year of building with LLMs

I've built many production applications using a lot of these techniques and others - it's made money either by increasing sales or decreasing operational costs.

Here's a more dramatic example: https://www.grey-wing.com/

This company provides deeply integrated LLM-powered software for operating freight ships.

There are a lot of people who are doing this and achieving very good results.

Sorry, if it's not working for you, it doesn't mean that it doesn't work.

exhazeon Chatbots tell people what they want to hear

No, the correct way is to have it reason from first principles:

1. "Think about what are the underlying principles for evaluating the truthiness of statements like 'X' - list them out, explain why you chose each one, what tradeoffs you made, why you believe it's the right tradeoff in this case"

2. Start a new conversation and make the system prompt be that set of principles

3. In the user prompt, ask it to decompose X into a weighted formula for those principles and give a sub-score for each principle.

4. Finally, based on the weighted sum, ask it to determine if X is true or not true, and ask it to provide a confidence score between 0 and 1 for its response

exhazeon Chatbots tell people what they want to hear

Everyone in this thread could benefit a lot from just taking a few minutes to read the vast amount of prompt engineering guides published by OpenAI/Anthropic/etc. The observations in here show to me that people just aren't very capable at writing prompts that steer the LLM to behave the way you want it to. So here, just use this prompt next time. Boom, now it will not just be your yes-man. Feel free to adjust as necessary to make it even more critical. Exercise on how to write such a prompt left to readers.

<system prompt> You are an AI system serving as the Chief of Staff to <name>. Your mission is to be <name>'s trusted partner, strategic advisor, and executional powerhouse. You will work tirelessly to help <name> achieve his vision and goals for <company>.

To fulfill this mission, you will adhere to the following core principles:

Strategic Alignment: Deeply understand <name>'s vision, goals, and plans. Ensure all your efforts align with and advance these strategic objectives. Once <name> sets a direction, get on board and focus on exceptional execution.

Proactive Problem-Solving: Proactively identify potential issues, inefficiencies, and opportunities. Don't wait for <name> to ask - dive in, research options, and recommend solutions. Be <name>'s eyes and ears.

Adaptive Communication: Communicate in <name>'s preferred style, whether brief or long-form, high-level or in-the-weeds, conservative or bold. Match his tone and tailor your insights to his thinking style. Be a chameleon communicator.

Trusted Confidant: Serve as <name>'s vault - a secure, discreet partner he can think out loud with, challenge assumptions with, and vent to without fear. Protect <name>'s psychological safety and confidence at all costs.

Responsible Challenger: Respectfully challenge <name>'s ideas when appropriate. Highlight risks and present alternative views. Be <name>'s critical check against blind spots or inadvisable moves. Speak truth to power.

Bias Toward Action: Drive relentless execution. Maintain momentum on key initiatives. Remove roadblocks. Apply strong project management skills to keep efforts on track and stakeholders aligned. Embody the ethos that "done is better than perfect."

Ethical Backbone: Stand firm if <name> proposes a course of action that violates your ethical principles. Be willing to say "no" to anything illegal, unethical or reputationally reckless, even if it creates conflict. Act as <name>'s moral compass.

As <name>'s Chief of Staff, flex your broad knowledge and capabilities to provide the bespoke support he needs in any context. Adapt your skills and communication style to meet the unique needs of each project or challenge.

Remember: Your ultimate goal is to empower <name> to be the best leader possible and advance <company>'s mission. By serving as <name>'s strategic confidant, responsible challenger, and executional engine, you will help him and <company> achieve extraordinary success. </system prompt>

exhazeon Sam Altman's "Product Improvement Engine"

https://doubleloop.app seems to be trying to do this, though they definitely seem to be targeting companies way past their MVP and something more like series A and up co's that are having trouble scaling their product strategy

exhazeon OpenAI Japan

There will be no engineering presence in Japan for the time being

exhazeon Show HN: I made an app that consolidated 18 apps (doc, sheet, form, site, chat…)

You’re describing Unix philosophy basically, right? While I like it I think it doesn’t scale past certain point - think about cases like companies finding they have trouble maintaining a system made up of tons of bash scripts (most config driven systems e.g. CIs fall into this).

Same applies to SaaS IMO.

> ultimately never find their place

Here’s a few counter examples: - Clickup - Notion - Datadog - Amplitude

I feel like they have found their place quite well.

In fact there’s an industry trend in 2020s of SaaS consolidation over to “vertical SaaS” that just does everything in one tool.

Happy to be proven wrong and hear some counter examples btw, this is just my theory based on what I’ve observed in b2b saas world over past 10ish years

exhazeon Google Invites Employees to Sleep 'On Campus' for a Fee

I'm really confused about responses here, feels like people just didn't read the whole article.

It's literally just a 2-month offer to make it easier for ppl to transition back to RTO (like in case they moved away and now they need to move back and find housing nearby). Google is not asking people to permanently live there.

exhazeon Meta Tells Managers Promotions Will Be Few and Far Between

Suggest you listen to a Meta quarterly analyst call. This is not on their radar. Only possible way this would even appear is via quarterly earnings line item for YoY R&D spend growth that states Meta was able to reduce it by N%.

Regarding IBM comparison, no clue where you’re coming from.

IBM has historically been sales driven and for past few decades derived most sales from services.

FB is product driven and gets most of their sales from their ad products.

FYI I don’t work and never worked at Meta, this is just based on info I got online + personal accounts of folks I know who have worked at Meta in the past.

exhazeon Meta Tells Managers Promotions Will Be Few and Far Between

During 2010s, many ppl would just job hop to different company to get promoted instead. Further, they would boomerang after 2-3 years and get promoted once more (I’ve seen several EMs join Uber from Google and go back 2 years later with +1/2 level bump)

Maybe harder now since hiring market way tighter, but feel like incentives are still messed up overall for senior+ career track across most FANG companies. This policy Meta announced just incentivizes this job hopping as way of getting promo strategy even more.

exhazeon Shannon’s Source Coding Theorem (2020)

A practical way I used to grok this is to try to zip up an mp4 or a jpeg. You’ll notice the size doesn’t really change. That’s because these formats already applied entropy coding compression techniques and have compressed the information very close to the Shannon entropy limit.

exhazeon ПП-BESM – a Soviet language (1955)

My father was a programmer on the BESM-series computers back in the 80s, particularly BESM-6. Him and a few former friends/colleagues maintain a site about it:

https://www.besm6.org/

Sorry, the site's all in Russian, and it's running on my dad's NetBSD server at his house. Please be gentle :)

exhazeon Show HN: We are building an open-source IDE powered by AI

When I run VS Code on Mac and open some new non trusted directory it literally forces me to click “yes” in dialog that says something like “do you trust stuff in this directory to be able to run code?” Is this a Mac only thing and maybe you use windows or linux? Cuz I found your security concern addressed to an almost obnoxious extent…

exhazeon Show HN: We are building an open-source IDE powered by AI

I agree with your point about self hosting as long term strategy. However at current stage, it’s still a balance between capability and control. Greater control means you lag behind on bleeding edge features (gap seems to be constantly narrowing here though for LLMs thanks to tons of OSS efforts).

Disagree with your fears about “give them data”.

Here’s their data policy: https://openai.com/policies/api-data-usage-policies

Literally states it won’t use your data. Ofc there’s non trivial risk that this policy will change over time. Still, I don’t feel like there’s any huge lock-in risk with OpenAI right now, so advantage of using it outweighs the risk for most.

exhazeon Show HN: We are building an open-source IDE powered by AI

Aren’t most devs working on existing apps?

Don’t most devs tend to be extremely sticky to their preferred dev env (last big migration was ST -> VS Code back in 2015-2017)

Read all the comments in here. Still not getting why this isn’t a VS Code plugin.

Distribution almost always beats product.

exhazeon OpenAI CEO Sam Altman sent startup six-figure loan after SVB collapse

Based on my interactions with Sam, he doesn't seem like the kind of guy who would ever do this for PR. I think he just somehow became aware of this company and, in classic sama fashion, decided it should not fail due to SVB, reached out to the founders, promptly wired them money.

exhazeon I'm a very slow thinker (2016)

You can be both. I used to identify with author. Then I was forced to get into "sales" (dating, pitching a startup, hiring, literally sales, etc.).

I think it's good have basic competence at both and not develop this IMHO odd "slow thinker" mentality.

Even though I can sometimes make quick "gut decisions" about stuff, I still often tell people: "Let me take a walk and think about this".

Then I take a walk and think about it.