manidoraisamy

The ones that are gone are comments/documentation in code, unit testing, design patterns, datamodel/db design. I am happy documentation is gone. But I miss design, especially db design. AI generates SQL/code with no regard for backward compatibility or future maintenance.

Not as much as US. Everyone from trump to Altman is betting on fusion. But China is more pragmatic by focusing on making fission resistant to supply chain shocks in uranium. Since they are fast follower, their plan might be to catch up, once fusion is viable for practical use.

Yes. We use Gemini with mcp which supports photos and PDFs natively.

Ideally, you want to start small and iterate. With Promptrepo, you can use versioning to compare model outputs across different datasets. In the test UI, we calculate confidence scores using @promptrepo/score [1], which parses OpenAI’s logprobs and shows field-level reliability. Fields with low confidence are highlighted in red, making it easy to catch signs of overfitting or data drift.

[1] https://github.com/ManiDoraisamy/promptrepo-score

Our target customers are product teams like ours. I’m the technical founder and was building a feature to extract product prices from unstructured text. But my co-founder, who understands all the pricing formats — is non-technical. I built Promptrepo, so he could finetune the model himself and improve accuracy without relying on me. My bet is that this same dynamic exists in many other product teams.

Thanks again for the feedback and recommendation. We’ll update our site with clearer examples and target use cases!

Finetuning is technical, but OpenAI hides things like layers, learning rates, and uses LoRA under the hood. You just upload examples - usually around 50.

But even that’s too much for most business users. Choosing input/output fields, generating and validating JSONL still feels like coding. That’s why we built PromptRepo: it turns finetuning into a guided, no-code workflow using spreadsheets.

Thanks for the clarification. OpenAI projects in ChatGPT are meant for end users to get personalized help using their own documents, inside the ChatGPT UI.

Promptrepo is for developers and product teams to build new AI-powered features in their product. It’s about creating custom models that run behind the scenes in apps, not just improving a personal chat experience.

So while OpenAI projects use RAG for better chats, Promptrepo helps teams build and deploy fine-tuned APIs that serve structured outputs like JSON, labels, or extracted fields to build your own AI powered product.

It depends on the use case. For many business workflows, where structured data is key - spreadsheets are already the source of truth. But for chat-based or unstructured tasks, a custom UI might make more sense.

Can you please clarify what you mean by "OpenAI projects"? Are you referring to the playground or the API for prompting or fine-tuning?

Yeah, just like Dropbox was a passthrough for aws s3.

Edit: Sorry about the snide comment. But if this ends up as a simple utility for finetuning, I would be happy with that too. Just want to share a tool that's been very useful in building ai features in our products.

We don’t have automated evals, latency, or cost comparisons yet. But, Promptrepo does offer versioning and lets you deploy the same model across providers for comparison. Automating these comparisons is definitely on our roadmap.

Exactly! Finetuning needs at least 10 examples to even work. That’s why Promptrepo begins with prompting and schema-based generation when teams have little or no data. As they gather more examples, it gradually shifts to fine-tuning. It’s the classic cold start problem and we’ve simplified it for product teams who want to launch quickly but improve accuracy over time.

Thanks! Yes, that's a popular request from other developers as well. That's why our basic plan is self managed plan (Bring your own OpenAI account) - https://promptrepo.com/finetune/pricing.html

We have one month free trial as well. Free free to ping me if you need more time.

Thanks for the heads up. Looks like a traffic spike from HN. Can you please check now?

Does it support Claude?

If you have a tight deadline, I would suggest going with what you know i.e. version control. You can launch it first, get traction and then change it later to blockchain.

This is a common problem in India. WhatsApp limited the number of forwards to 7 per day, after some WhatsApp rumors led to murders. This sounds like a good solution for receivers. Question is, how do you stop people from forwarding?

Join startup school which starts on 22nd [1]. YC is going to have co-founder matching part of the course.

[1] https://www.startupschool.org

That's a great example of usability! Developers like us are used to libraries/frameworks and forget most algorithms and architecture (literally the word itself) come from real world examples. We expect the same in usability - get bunch of CSS/UI library and complete the job. This book is useful, if we want to learn about usability from real world examples, rather than cookie cutter CSS/libraries.

It is quite common. Don't go by the contractual obligation. If it is a small amount for you and your customer is going to stick to you, you might want to make him a happy customer by giving discount.

Meet the makers - https://www.3dprintedhouse.nl/en/

My original title for this submission was "Moderate level of egg consumption (<1 egg/day) lowers risk of heart disease". It was edited probably because it was clickbait-y? [1] Nevertheless, I feel that the summary was quite surprising - "Daily consumers also had an 18% lower risk of CVD death and a 28% lower risk of haemorrhagic stroke death compared to non-consumers."

[1] An egg a day might reduce your risk of heart disease, study says - https://edition.cnn.com/2018/05/21/health/eggs-heart-disease...

If you are a developer, you can try these:

- Be active in Q&A sessions and community events. They might also have an expert community that you can join [1]

- Participate in alpha programs and connect with the product managers

- If your app is widely adopted, offer them to do a customer success story about unique iOS feature that you might have used. If you reach that stage, they are unlikely to ban your app without warning.

[1] For Android - https://developers.google.com/experts/all/technology/android

I'm sorry for your loss. Do you know anyone in the AppStore team, whom you have talked to during the good times? If not, do you know any fellow developers who might have such contact? Your best option is to tell your side of your story (your efforts/dependence) to a person, instead of assuming the official communications as the final word.

EDIT: In future, I suggest you invest time in establishing relationship with these teams (like going to WWDC), especially when you are betting your entire revenue on such marketplaces.

This is pretty good already. You might want to consider narrowing it down further to "Analytics for single page web apps" like you mentioned above. It might make it more understandable for nontechnical folks, even if it doesn't cover the full AJAX scope.

BigQuery is great for read and analytics. It's one of the best products I have used. But, it may not suit high frequency inserts (not a transactional DB). Also, the insert only "limitation" may not exist in the future.

I don't think it's as simple as that. This is a data mining exercise - to read our customer's subscribers data and to move from a "dumb pipe" to a "personalization engine". From mailchimp CEO's blog: "We came to a fork in the road, and choosing the “personalized transactional” path with Mandrill suits us and our customers better than the “utility” path."

http://blog.mailchimp.com/important-changes-to-mandrill/

"your Subscribers' email address, credit card information, purchase history" - These are super sensitive! It takes more than 3 months of legal process and convincing to get our customers to share these under strict data privacy clause. And here comes mailchimp and want to sniff it under the hood in the name of migration. They will literally land us in jail, if we don't migrate.

There is something more serious than migration and price increase. By migrating to mailchimp you are agreeing to mailchimp's terms and service. If I understand it correctly, now they can read all your customers' email address and content (presumably for machine learning & send optimization). This is in direct conflict with the agreement we signed with our customers on data sharing. Unless I am mistaken, this seems to be deliberately omitted in mandrill communication to mislead and collect data to create "network effect" from our customer data.

Perfect! This is one of the best documentation I have seen recently!