jmathai

This may not be a popular opinion but I’ll say it anyways. I’ve done a lot of this and am sharing anecdotal information.

The claims that AI writes inefficient code may often be right. If and when it’s an observed problem, you can use AI to improve performance.

The claim that AI writes insecure code is probably also correct. You’ll want better safeguards here based on your product. The infrastructure you choose to deploy on can help as well - the LLM can help you design for this.

The claims that AI will delete your database are unlikely and an indication that permissions were too loose. You can also sandbox the code which runs. Again, AI can help you design this.

This is not an endorsement to blindly deploy code you haven’t verified.

My feedback is to be as optimistic as possible. Learn where the edges are. Find ways to patch them systematically. And let your confidence grow or shrink from your own experience.

I use Claude Code the way others use Lovable. I'm comfortable with the terminal and Claude Code but I think Anthropic (et al.) is closing the gap rapidly. And Claude Code + Claude Design is way more capable than what I think Lovable is.

I assume Lovable had to build around constraints to reach non-technical people at the time. But the models and tooling have improved such that those constraints are melting away. I'm not sure what advantage Lovable has in the long run.

I don’t think end users will purchases subscriptions out of their own pocket for work en masse.

If the value is there, companies will.

Funny that it bashes a future where AI destroys jobs but embraces one where it destroys relationships and wellbeing.

I avoid subscriptions like the plague and I was getting enough value from Claude that I was paying for the Max plan for a couple months. I don't know how much Anthropic needs to make to be sustainable but I think it's worth north of $50 or $100 per month for many knowledge workers.

I’m exploring how to use AI for faith oriented experiences and products. There are two specific areas I’m interested in. I’m furthest along with this app in the App Store. https://trysojourn.app

1) grounding LLM scripture references in the actual text and eliminating hallucinations. This is the platform for that. https://withlattice.com

2) how to use AI to get people into scripture more as opposed to being a substitute for it. This is the product I made for it - it uses #1 above. https://trysojourn.app

I have Claude implementing features from a spec, running tests, creating a build, and then publishing to TestFlight. All I have to do is verify that the build works once I get the TestFlight notification.

I have found a key is to use end-to-end tests and not unit tests.

This has been working well so far after around 100 pull requests for an app I recently decided to make. So far it’s been good.

It’s free, no subscription.

[1] https://trysojourn.app

I’ve done similar with backend projects utilizing GitHub actions to run tests and publish to staging for verification.

The cost per token is super low. If you're used to paying OpenAI or Anthropic API-based fees then the same workload on DeepSeek feels free.

On DeepSeek's platform page yesterday I saw that prices during peak hours would be doubled (IIRC). Today, it's just a generic message that prices will go up significantly.

If you want to make money, work in big tech. If you have an insatiable itch to build something (product, business), start a company or startup.

Took a look at StepBible. Looks really cool. My email is in my profile if you wanted to connect.

I tried using Deepseek with Claude Code and it under performed. The use I was referring to was within a product and I feel it's a lot of value for the cost.

https://api-docs.deepseek.com/quick_start/agent_integrations...

That’s okay, IMO. Ultimately it’s the readers responsibility to apply their own intelligence. But the app should not misquote scripture itself.

It’s pretty polished but you’re asking the wrong person. Feedback has been positive so far :).

Philosophically, I don’t believe we should outsource the accuracy of scripture to any single entity (let alone a for profit secular one).

So it’s less about model choice and more about governance of scripture.

I will check out the link you sent for sure!

Very true. The verses selected come from a tool call. The LLM queries the app for relevant verses. It picks the keywords - so perhaps there’s bias there but the verses are handed to the LLM.

But there are ways to control and constrain the LLMs and what the user is presented with.

These are all top of mind for me and why I felt there could be a better option than asking ChatGPT directly.

You’re correct. The hard line for me is ensuring that any scripture presented to the user is verified correct. LLMs can’t be trusted in this regard.

One feature of the app is that all scripture is verified and what’s show to the user doesn’t come from the LLM at all and instead a trusted source.

I think exploring scripture this way does not alleviate you from struggling to learn and apply it. It hasn’t for me.

Huge difference while both are competing. But it’s a huge risk in a market with such high cost of entering. If either divests then it rapidly degrades.

I’ve been using v4 flash for an app I’m building [1] and it’s amazing how cost effective and good it is coming from having always used gpt, opus and sonnet models.

It’s so cost effective I can offer a generous free tier since my goal isn’t to make money with it.

[1] https://trysojourn.app

I don’t disagree with what you’re saying but I said before any of this mess, and just dealing with the ticket purchasing experience, that this would be the last World Cup I attend.

It’s sad though. The World Cup, while imperfect, remains an event that brings the world together. We don’t have many of those left.

The experience interacting with FIFA is so terrible already. I can only imagine how much worse it would be if it turned into a profit machine.

Tobacco is a good example. Legislation that you could not smoke inside or nearby buildings did a lot to curb usage.

What’s challenging now is that corporate interest in government seems to have increased rapidly. It’s unclear if legislation is viable.

I agree with everything the article says. I think there are two questions that we should be asking.

1. Can we point to a time when Capitalism peaked in terms of a balance of maximal benefit for as many people as possible? In my lifetime, it was somewhere in the late 1990s and early 2000s. No singular answer but something better exists than where we are now.

2. How do we rewind capitalism to that point?

They employ 3-4 people to man their drive through. No wonder it's such a great experience. Subconsciously, I think I pick CFA over alternatives, in part, to not have to deal with poor customer service.

Odd. I receive a non English website when clicking on that link.

I understand that technology has advanced us in many ways. I also think it started to regress us in other ways. At rates where the regression feels faster than the progression.

I think many of us feel this. Even if it is expressed differently.

I feel your pain points.

> The second problem is it can't just work with the file system structure and organization

I presume you know about storage templates and ruled them out.

Oh wow. Thank you and great to run into folks from that community.

I’ve preserved all of the tags and albums from my Trovebox library. Had high hopes of integrating it into Western Digital NAS devices but…big co :/

I don’t trust Immich myself :). Its read only external library feature is what made it an option for me.

Given this is a largely technical crowd, I feel it my duty to share just how good (and free/open) Immich is.

If you’re like me and don’t want to be an “admin for life” then it’s still for you.

What has worked for me for over a decade is to keep the source of my photos in a boring old folder (backed up to my synology and Dropbox). And then layer photo viewing and sharing apps on top.

The day I’m sick of Immich and there’s a better alternative, I switch.

I’ve written about how it works as I’ve gone along. Recommend reading and putting your own twist on it.

https://jaisenmathai.com/articles/my-ridiculously-robust-pho...

https://medium.com/vantage/understanding-my-need-for-an-auto...