rishabhpoddar

I was actually working on something like a "claim checker" recently. Inspired by stuff like Codeql. But I realised that all the really valuable business facing claims can't be checked just by static code analysis, and need to be done via integration and unit tests only. Given that claude code can pretty much write all the runtime tests fairly easily and quickly, I don't see how anything else can be better.

Fairly interesting! A few questions:

- When asking the user for an input, does it support multi turn user conversation?

- Who is this made for? Developers would much rather just use python directly with coding agents; non developers would probably still be afraid of seeing the python-ish syntax in the md file + they wont be able to use existing coding agents cause there would be no examples of this new language in any of the LLM's training data.

But it's an awesome attempt! I like the idea overall, and the design of the website!

What if my network is not on the app?

For most people, they never get any feedback from these platforms anymore.. there are just SO MANY PH and HN launches everyday.. that unless you pay some big PH hunter, or are a part of YC (for HN launch help), your post will just get ignored.

It really depends on your product, but cold emails / DMs still works if you try enough times. Launching on product hunt, HN at least gets some eyeballs and maybe some of those eyeballs convert to people who try your product.

Does anyone know if it won't leak information across slack channel since it has the memory feature?

I ask it to write unit test with close to 100% coverage. This has worked well for me so far

It will always be important as a networking / social experience with very like minded folks. But for learning the subject, not so sure

As models get better and better at following instructions, don't you think this will become lesser and lesser useful?

How is this better than higgsfield?

HOw does this compare to something like this: https://teamcopilot.ai/blog/automatic-blog-workflow, specifically in terms of the blog post creation flow.

The problem is, how do you know if someone is an expert in the first place? A LOT of podcasts are done with random people who claim to be experts, spreading misinformation. How does your product filter those out?

The only way to validate an idea is to build an MVP fast, launch it, and see what people say (if anything at all).

I was literally thinking of something similar, but more on the lines of somehow subsidising costs for using claude code / codex if you choose to share the session trace.

oh yea! three body problem is awesome! But it already has a TV show for it which i had watched

thanks!

I agree that CORS is hard to understand and fix. I was the CTO at an auth company and SO many of our users used to run into various CORS issues and asked questions on our support. However, I'd now argue that developers don't need to understand CORS anymore.. cause claude / gpt does! Just throw in the error in claude code / codex and it would fix it.

Do i have to put in what I ate throughout the day? How accurate / easy is it for me to put in each ingredient that I ate and its weight? Instead, can i put in a weekly meal plan that I already have?

I use a mix of claude code, codex and cursor (composer 2.5). Whichever has more credits left.

Seems interesting, but it's really hard to understand what it's needed for. Could you elaborate?

We do need code, cause humans can't read and understand binary. Let's say, even if we do get perfect English -> Binary, often times what people specify in english can be vague or just not what they actually meant. So if the AI just outputs binary, it will get very difficult to be sure that what is implemented is actually what the user meant.

- Asking google / gpt - searching for "books like ...", and then seeing goodreads review. - Often I see people on X post about books, so I add it to my "want to read" section in goodreads. At this point i have nearly a 100 in there, so good to go for 2-3 years.

I just use codex / claude code to do reviews. It does a good job, and it's easy to explain and navigate.

Or, I can just tell claude code to make it for me in the same timeframe?

Can this do backtesting?

I feel like depending on what you are making, distribution moat has always been stronger than coding moat. Even before AI, crappy, well distributed software used to beat good, but badly distributed software.

Long ago I had learnt Haskell, and when I discovered point free programming with it, i got addicted to making everything point free. It was totally useless in terms of performance, made it harder to do read the code, BUT, I was obsessed nonetheless.