Show HN: Create your own finetuned AI model using Google Sheets

[QueensGambit]

Hello HN,

We built Promptrepo to make finetuning accessible to product teams — not just ML engineers. Last week, OpenAI’s CPO shared how they use fine-tuning for everything from customer support to deep research, and called it the future for serious AI teams. Yet most teams I know still rely on prompting, because fine-tuning is too technical, while the people who have the training data (product managers and domain experts) are often non-technical. With Promptrepo, they can now:

- Add training examples in Google Sheets

- Click a button to train

- Deploy and test instantly

- Use OpenAI, Claude, Gemini or Llama models

We’ve used this internally for years to power AI workflows in our products (Formfacade, Formesign, Neartail), and we're now opening it up to others. Would love your feedback and happy to answer any questions!

---

Try it free - https://promptrepo.com/finetune

Demo video - https://www.youtube.com/watch?v=e1CTin1bD0w

Why we built it - https://guesswork.co/support/post/fine-tuning-is-the-future-...

[Centigonal]

Your biggest obstacle is proving fine-tuning is more effective than prompting, workflow design, RAG, etc during the initial pass. Most of my customers are still getting big improvements by picking the low-hanging fruit with those approaches. A much smaller fraction is at a place where they're ready to start fine-tuning. Obviously, this will change as AI programs mature.

[manidoraisamy]

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.

[polskibus]

Can you share an example of such real world win where fine tuning was less effective ? I’m curious about sample business cases.

[rgbrgb]

Incredibly crowded space, but this is a great insight and UI... engineers probably have to integrate the model but we should empower non-technical customer-facing people to give feedback to the model in a way that improves it.

The blocker for me (and likely other cost-conscious early stage groups)? I have free credit and existing integrations with more mainstream platforms (OpenAI, anthropic, together). Trying this out will cost both eng time and money, so I won't be an early adopter. I wonder if there's a way to pass the cost through / use my API keys with credits. Maybe it's more for enterprise teams or cases where you're already confident about the fine-tuning approach.

Anyway, congrats on the launch!

[manidoraisamy]

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.

[ygreif]

I love integrating into spreadsheets. Super easy to use. Reminds me a bit of mailmerge.

I don't have much experience with modern finetuning, but isn't it highly technical. How many layers do you want to change? What is the learning rate? Does that need to be visible to the user? How many examples are needed in practice

[manidoraisamy]

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.

[ipaddr]

I don't like the pricing. $38 a month headline much smaller billed annually with the price more than doubling. This is the trick your customers strategy and hope for no chargebacks.

[QueensGambit]

I understand your concern. We’ve had cases where short-term users abused our product for phishing, which led us to remove the monthly plan initially. To address that without locking out genuine users, we’ve added an option to extend the trial for up to 3 months before choosing a monthly or yearly plan. Feel free to message me if you’d like more time.

[scosman]

What’s the thinking of spreadsheet first? Just making it super accessible for people who already have data?

I’m building a UI for fine tuning (and evals, and synthetic data gen) - https://github.com/Kiln-AI/Kiln - and went the custom UI route. From chatting with folks - most people don’t have datasets, and need help building them.

[manidoraisamy]

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.

[heresjohnny]

Neat! Wonder though if you should be even offering the BYO option as a separate lite package. As a dev I would not buy this and as a non-tech person I would be confused by your pricing page.

But I do see the value! Think sales or marketing folks looking to get a bit more hands on. These will likely be your first visitor and be okay with your 50 dollar price. Then, their IT department will say “we want to hook up our own API key for that,” to which you can confidently say “sure, we can do that too.”

N=1, just my two cents. Good luck!

[raylad]

Who is your target customer?

From reading the site, it's not clear that someone who doesn't already know about fine tuning and want to do it would know what your service does or why they would need it.

Recommendation: describe the process, and give some examples of applications.

[manidoraisamy]

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!

[nbbaier]

Heads up I get "ERR_CONNECTION_RESET" when navigating to https://promptrepo.com/finetune

[manidoraisamy]

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

[ricktdotorg]

this is great, and right up my alley. i built a customer support chatbot using a google sheet for our CS folks to input & structure their question/answer pairs. that is auto xformed into markdown and fed into the bot for context. it works fairly well considering how simple it was to do. i'm really intrigued as to what promptrepo can add to that. will definitely give it some R&D time!

[QueensGambit]

Yeah, Google Sheets is a surprisingly powerful interface for business teams, especially when they’re the ones curating the data. Curious to see how Promptrepo fits into your workflow. Happy to help if you explore finetuning on top of your existing setup.

[ivape]

This is awesome, I don't have a fine-tuning use case yet, but I can't imagine something being easier than a spreadsheet.

[rallyvite]

Agree, not super familiar with this space but that you are using GS as the UI makes this super accessible to so many more people given familiarity and low intimidation factor. Best of luck.

[sneha_tamal]

How do you make sure the fine-tuning process stays effective without the risk of overfitting on specific data?

[manidoraisamy]

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

[littlestymaar]

Why Google Sheet though? Why would you want your customers to give their training data to Google?!

[QueensGambit]

Don’t want to sound like a shill, but Google doesn’t use data from your Google Sheets to train its models or for advertising. By default, your data stays private and is protected under Google Workspace’s privacy policies.

[justanotheratom]

FineTuning Ops platform requirement - compare evals, latency, cost across models.

[manidoraisamy]

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.

[gitroom]

honestly i like seeing tools get simpler like this, makes me wonder though how much of the real value comes from the tech itself vs just making everything less scary for folks who aren't engineers. you think easy interfaces actually help people get better results or just get more folks trying stuff?

[QueensGambit]

Depends on who’s using the tool. Developers might be fine coding a form into their app, but an HR person needs a form builder. Similarly, the data for training models usually lives with domain experts, but they don’t have a tool to actually do the training. That’s why, in this use case, a simple interface makes sense, IMO.

[labrador]

How is this different than OpenAI projects?

[manidoraisamy]

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

[mogili]

Essentially this is a frontend to automate the process of converting a csv file into jsonl and pass through a fine-tuning service.

[manidoraisamy]

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.