Ask HN: What makes your AI special that can't be replicated by competitors? | Hacker News Reader

Ask HN: What makes your AI special that can't be replicated by competitors?

5 pointssaliagato10 comments

Took the idea from a comment [1], I thought it was worth opening a new thread for discussion.

[1] https://news.ycombinator.com/item?id=36943615

It's trained on my data which is perfectly suited to my problem, which my competitors don't have.

Cool. Do you mind me asking what do you mean by “trained”? Fine-tuned on your data?

I think the specificity of data is general, but I have a number of models that predict preferences and engagement for myself and communities. The production models either use classical ML with bag-of-words and other features or classical ML on BERT-like embeddings. The data is a set of judgements about individual documents.

I tried fancier “fine-tuning” methods and the noisy nature of the data means they won’t perform vastly better than my current models and have training times more like 30 minutes instead of 30 seconds but it is worse than that because if I want good results I might need to fine-tune 8 models and pick the best.

So I stick with my system because I can retrain as much as I like, run thousands of experiments a day if I want, etc.

This is the answer. Very smart people [1] are disregarding the worth of proprietary data.

[1] https://twitter.com/nikiscevak/status/1650295590809636864

+1 this

I can't wait till the tech becomes even more easy to use, so one could work on ML projects with the same ease I had in the late 90s when learning from inspecting HTML source of pages I liked.

(also, curiously most of the AI services I'm paying for are also trained on my data!)

I believe your question is asked in good faith, but it seems important to bring up the well-written-about Startup vs. Incumbent concept that's been all over startup culture for a long time now. Meaning, just because there's an incumbent in a nearby space that _could_ replace you if they wanted, doesn't mean they will, and that reality shouldn't necessarily prevent a person from venturing into the creation of that idea. So, sure, Google _could_ build anything that most of us could build, much faster, purely because of their available resources.

That said, the answer to a lot of AI questions including my own is "because my data is better than my competitors".

We've created TranscribeMe: https://transcribeme.app

A bot that transcribes WhatsApp and Telegram voice notes to text; if it's a looong voice note, the AI sums it up! It has also ChatGPT integrated

Developed by Rather Labs: https://www.ratherlabs.com

To make a good use of it is already a big advantage currently. One individual creating a new tool based on the current AI tech moves so much faster than a corporation or even a startup nowadays that speed is what competition can't replicate.

One indie developer only needs 1/100 of the success he would achieve with a startup. And a few will achieve same level of success. The genie is out of the bottle.

Its integration with public and private data along with layers of automation allowing its use way beyond chat bots. From individual tool to something that can power large parts of a business.

https://flowch.ai - Currently free to use.

I found problems hard level and obfuscate my code, also hide some steps data of preprocessing on server side, so only wizards can repeat the magic.

Reply on news.ycombinator.com