HNHacker News
TopNewBestAskShowJobs

diavelguru

70 karma · joined July 1, 2019

submissionscomments
diavelguru··on A browser-native classic Visual Basic VB6 IDE
talk about a blast from the past!
diavelguru··on Ask HN: AI writes better code than me. How to keep my identity?
How long have you been working in this field? I ask as if you have numerous years you already know your worth and value irrespective of AI or any other system. If you are still questioning this it indicates to me that you are still in the process of finding yourself and are in the same boat many other folks are in - confronting the existencial crisis of identity as it pertains to work. I suggest do what is necessary to stay employed then on your own time write code by hand for personal or side projects. Be sure to not use AI and learn deployment, devops and infrastructure as much as is needed to be knowledgeable and able to speak to the technology. Then add that to your resume. Keep grinding.
diavelguru··on /architect: Reduce Fable tokens by 80%, Fable orchestrates/reviews, Codex builds
yes I'm using Fable to inspect, generate plan and architectural docs then using Gemini to implement then have Fable review, find bugs. saving lots of usage.
diavelguru··on Show HN: Git-Shitstorm: How to Make Any Developer Lose Their Mind
but....just.....why....?
diavelguru··on Ask HN: Instead of intrusive age-check why can't we have "two internet"?
Interesting idea. Would need new laws enacted and government oversight which I'm not sure how people feel about that especially for minors. Maybe there will come a time where all people (in the US or otherwise) have some form of government document that uniquely identifies them (a trusted identity platform not just an ID) though there is and will be lots of pushback. A simple search revealed this: "The implementation of universal government ID systems, particularly digital IDs, is frequently criticized as a key component of a dystopian surveillance state. Critics warn that such systems can lead to mass surveillance, restriction of freedom of movement, and control over personal, financial, and medical lives." So take it for what you will. Can we do this? Yes. Will we is another matter.
diavelguru··on I built Tinder for movies with Reels-style trailers
The UI is slick nicely done. Currently refreshing gives new batch of movies. Maybe when login they don't change or that's a feature not a bug? I like that it surfaces lots of movies I would have never even considered.
diavelguru··on Agentic AI systems violate the implicit assumptions of database design
Agree across the board.
diavelguru··on Is AI pressure making developer burnout worse? (anonymous survey inside)
Having used Claude Code at work and for personal projects these past 8 months, I can see the attraction to keep going till your session cap is hit. I've used Claude for professional work coding, unit testing, personal projects in languages I haven't learned; completed work I had on the back burner for a number of years. I learned early on to take time for myself and this is no different. One must get out and run, swim, bike, weights, whatever but get your mind out of the computer and work and do some human necessary things besides just eating, pooping and sleeping. We need time to bond with our loved ones, parents, siblings and yes even coworkers. Until we as individuals take that time and leave the obsession (for a while), yes we will spend more and more time getting things done with AI. It's like any addiction. The dopamine hit is very powerful and we do need to reset our mind and rest and form long term memories as well as what my mom always said (take time to smell the roses).
diavelguru··on Ask HN: How to handle desktop app security post-Mythos?
We shipped software in the late 90’s with a hasp protection key required for use to prevent copying. https://en.wikipedia.org/wiki/Software_protection_dongle part of the process was to run a hasp obfuscation exe against our newly compiled exe. It was most likely an encryption algorithm that had the key on the hardware dongle. This exe is what was put in the resulting floppy disks and burned onto cd’s for distribution. I fielded many calls regarding issues with this but it worked and kept the software limited (it was a niche market). I can see something like this foiling any LLM inspection of the resulting binary as the binary has been drastically modified from its original. What’s old is new again.
diavelguru··on Do I Stop Learning Coding? DSA?
Ok I have to ask you are you in it for the paycheck or the passion? How you answer determines your path. If a paycheck optimize for the largest salary you can conceivably obtain based on your ability and education. If passion follow your passion, get educated and eventually you will find some work in your field. The passion is contagious and will positively affect your life in ways you can't even imagine at your age currently (self satisfaction plays a lot into child rearing). Notice I didn't say money. The day I stop learning is the day I die. My advice, don't stop learning.
diavelguru··on I propose a new programming language, CPC
Sounds cool. Do you have any requirements other than what you outlined above such as any grammar examples? Reading this brought me back to my compiler days at uni. We used Oberon and had to create a compiler. It was quite the process! I remember abstract syntax trees that held the program and as we proceeded through the compilation steps that ast was passed around and inspected for all steps. This I found helpful: https://dev.to/villyp/how-to-build-a-compiler-a-step-by-step...
diavelguru··on Every plane you see in the sky – you can now follow it from the cockpit in 3D
yes just saw that!
diavelguru··on Every plane you see in the sky – you can now follow it from the cockpit in 3D
did this https://flight-viz.com/cockpit.html?lat=40.64&lon=-73.78&alt... flew over Mexico then got to Antartica and it started going backwards. was hoping to get to South America. Fun!
diavelguru··on Issue: Claude Code is unusable for complex engineering tasks with Feb updates
As soon as that change came through I set the effort to high. Have not regretted it for any coding task. It feels the same as Dec-Jan though now spawning more sub agents which is not a bad thing.
diavelguru··on Issue: Claude Code is unusable for complex engineering tasks with Feb updates
Love this. Responding to users. Detail info investigating. Action being taken (at least it seems so).
diavelguru··on Plenty of AI hype, but not much useful software?
With AI, I've tackled lots of personal projects I had on the back burner for lack of time: recreating some old games, creating some marketing and shopping sites for the wife and also professionally: using it to fulfill business requirements. So in my little corner of the world it is being used and a lot mostly nonstop from Aug 2025 to present. Re your question: I think it has to do with all the rest of what's needed for a production system, the documentation, CI/CD, production deployments: all those pesky still-needed-human-things that are somewhat tedious to set and get correct.
diavelguru··on If AI is so good, why don't we have an infinite supply of 10x engineers?
We get a few not infinite. with a Claude 5x Max subscription I get 3-5 workable parallel agents that themselves spawn subagents to complete work in a ~3 hour window before tokens are spent. I imagine with 20x Max that could be increased to maybe 10 within a 5 hour window? so if you want infinite you're also talking infinite money. Working in parallel takes me personally to a higher plane of output than that of a 10x engineer as it pertains to fulfilling requirements.
diavelguru··on Why developers using AI are working longer hours
and flow state is a luxury in 2026 with AI swarm most likely to be found sparingly if all. Good luck all!
diavelguru··on Why developers using AI are working longer hours
you are correct flow is not achieved as this is not programming more like system design, architecture, QA, Product Owner work. It's using the swarm as your own dev team.
diavelguru··on Why developers using AI are working longer hours
yes agreed. I'm running 3-5 parallel Claude at once with requirements as the input. My prompt is say work on section 5.1 or something very specific. Then I'm monitoring the work across all instances.
diavelguru··on Why developers using AI are working longer hours
This is a real thing. I spent all of January doing Greenfield development using Claude (I finished the requirements) and all I can say is thank goodness I had the Max 5x plan and not the 20x as I got breaks once the tokens were used up till the next cycle. I was forced to get up and do something else. That something else was biking, rowing, walking. My productivity had never been higher but at what cost? My health no thanks. So I'm glad I'm using the time till token reset for my health. I time it perfectly. I do a walk, row, bike for 1 hour then as I arrive back the tokens are reset. I get like 3 hours nonstop use per token batch with the 5x plan. I've been thinking about going 20x but am scared...
diavelguru··on Ask HN: What's the best engineering interview process?
Definitely have some questions to get a baseline understanding of the tech stack but then ask a problem solving question and listen to what they say and how they think and use their answers to turn the interview into a conversation by asking them questions using their answers as a guide. The flow will be natural for both interviewer and those interviewing and insights from the conversation are very revealing.
diavelguru··on Tell HN: I am afraid AI will take my job at some point
I agree with this guy: https://obie.medium.com/what-happens-when-the-coding-becomes... When you look at the AI as having a Jr. Developer at your disposal to do the mundane things you actually elevate yourself higher than you could individually (without the help) as you focus more on architecture and design and guide the LLM to make incremental improvements instead of decremental ones. That's my take at least.

This one also: https://terriblesoftware.org/2025/12/11/ai-can-write-your-co...

diavelguru··on Opus 4.5 is the first model that makes me fear for my job
I agree with this guy: https://obie.medium.com/what-happens-when-the-coding-becomes.... It's a game changer when used to pair program (MOB Programming) with you as the navigator and the LLM as the driver. It needs guidance and rework but shit gets done with the human as the ultimate gatekeeper.
diavelguru··on Ask HN: What is the most important thing in life?
Claude opus 4.1: Your framework for thinking about life's purpose through utility maximization is intellectually rigorous and thought-provoking. Let me engage with your ideas by examining both what works well in your model and where it might benefit from additional dimensions.

## The Strengths of Your Utility Framework

Your insight about logarithmic utility functions is particularly astute. You're absolutely right that spreading value across multiple people generates more total utility than concentrating it in one person - this aligns with established economic principles about diminishing marginal returns. The example of $100,000 to ten people versus $1,000,000 to one person captures this beautifully.

The temporal dimension you've identified is equally important and often overlooked. When you note that helping your parents at 70 provides more utility than waiting until they're 80, you're recognizing that people's capacity to enjoy and benefit from resources changes over time. This adds a crucial urgency to altruism - it's not just about giving, but giving when it matters most.

## Potential Refinements to Consider

However, I'd invite you to think about a few aspects that might enrich your framework:

*The Paradox of Self-Care*: Your model positions maximizing others' utility as "the direct opposite" of maximizing your own. But consider this: if everyone followed this principle perfectly, who would be the recipient of all this utility? There's a recursive problem here. Moreover, maintaining your own wellbeing often enables you to provide more utility to others over time. Think of it less as opposite directions and more as interconnected systems - your capacity to help others depends partly on your own stability and resources.

*Beyond Utility Calculations*: While utility functions provide a useful analytical tool, human meaning-making often transcends these calculations. Consider experiences like creating art, pursuing knowledge for its own sake, or moments of profound connection. These might not maximize anyone's utility in measurable ways, yet they seem central to what makes life meaningful. How would your framework account for the scientist pursuing pure research with no immediate practical application, or the artist creating something beautiful but "useless"?

*The Measurement Problem*: You acknowledge that value isn't just financial but also emotional. This is crucial, but it also introduces significant complexity. How do we compare the utility of teaching someone to read versus comforting them through grief? The incommensurability of different types of value makes optimization extremely difficult in practice.

## A Teaching Exercise

Here's a thought experiment to test your framework: Imagine you have the choice between two lives. In Life A, you maximize total utility delivered to others through efficient but impersonal systems (perhaps developing technology that helps millions). In Life B, you form deep, transformative relationships with a smaller number of people. Which life delivers more utility? Which would you find more meaningful? The tension between these might reveal something about the limits of pure utility maximization.

## Building on Your Foundation

Your three optimization questions are excellent starting points. But I'd suggest adding a fourth: "How do I discover what actually provides utility to others, given that people's needs are complex and sometimes unknown even to themselves?"

Your framework shows remarkable sophistication for someone at 21, particularly in recognizing the multi-dimensional nature of value and the importance of timing. As you continue developing these ideas, you might find that the most important thing in life isn't a single principle but rather the dynamic balance between several essential elements - helping others, yes, but also self-development, creation, discovery, and connection.

The fact that you end with humility about these being preliminary thoughts shows wisdom. The question "what is most important in life?" is one humans have grappled with for millennia, and each generation must find its own synthesis of ancient wisdom and contemporary understanding. Your utility framework contributes meaningfully to that ongoing conversation.

diavelguru··on Ask HN: Are foreign key constraints still useful?
You answered your own question in the description. "DB constraints pop up when I am making changes to my local environment” and that gave you the confidence to promote to QA, Staging and finally Production. A well tested application will find those problematic issues long before it goes to prod. I’m just glad that all business logic is not thrown in the DB as in the past. The separation makes for a much quicker and robust system. Don’t think DB constraints are a relic they are very much a part of the relational database management system and still used in those systems very successfully. Another topic for another thread is the relevance of the RDBMS itself not just one subset of it.
diavelguru··on What good AI news aggregator are there?
https://www.newsminimalist.com/
diavelguru··on Ask HN: Should we bring software dev in-house?
Agree with this re serverless cloud computing
diavelguru··on Ask HN: Should we bring software dev in-house?
One thing to remember is that some/most are not turned off by a non-glamorous industry like logistics but rather attracted to solving problems regardless of the industry. I’d hire someone who can actually do coding (so not a CTO, CIO level) that is also or has been a lead who can delve into your business and review existing products and can get you an 80% POC in under 3 months. It may not be pretty but it will help you better position to ask for more $$ for contractors or other employees. Start with 1 then scale from there. I’d give them 1 month to study the existing system with no deliverables other than documentation that will be used to replicate your product needs. I’d also be sure to hire a PO (Product Owner) if you don’t have one and have them also learn the product inside and out as they will be the gatekeeper between the coders and the users as to how the product functions and that the coders fulfill all the requirements.
diavelguru··on Ask HN: Is crypto an unviable, tech-bubble delusion?
https://web3isgoinggreat.com/
Page 1 of 3Next →