mrtsepelev

6 ms·

1.
by mrtsepelev 24d ago
What harness would you recommend? I’ve tried Pi but the model struggled to stay on track after the compaction. I have only 48gb of ram, so can fit only 80k context max, so good compaction is must.
2.
by mrtsepelev 1y ago
While not a plugin, I've had a great experience using the free tier of Gemini CLI over my md-repository. It's rarely run out of context, unlike with code repositories, and it's super easy to feed relevant context just by ment
3.
by mrtsepelev 10y ago
We had some tests setup and run (some of them were a clear success, some -- not that much). And we're currently working with one huge Russian company, but still could not publish a case yet (but surely will as soon as possible). We kno
4.
by mrtsepelev 10y ago
Reasonable concern. We have a basic FAQ, but it does not say anything about integration (which is actually pretty simple and require only to add our js script). Will try to fix it.
5.
by mrtsepelev 10y ago
Sure. Ofk we used PH & HN and Reddit in the beginning (and many others startup platforms). Tried to reach industry bloggers (had some effect) and reporters (did not succeed here). Tried a content (which brought high quality leads, but q
6.
by mrtsepelev 10y ago
And what do you mean by "market it a little differently". We played with positioning/wording for a few months and while it affects our registration numbers, it did not help to activate users. Maybe we did it all wrong - we tr
7.
by mrtsepelev 10y ago
The scheme is simple: the more -- the better. Though, you should not be put back by the false assumption of your traffic flow being too weak -- give it a try and see how it goes. Usually it requires about 10k unique visitors per month to st
8.
Ask HN: We built a product that nobody wants. What’s next?
12 points by mrtsepelev 10y ago|15 comments
9.
by mrtsepelev 10y ago
Actually, it is. We're currently running simple personalization campaign with two versions of landing pages which has different design, messaging, etc.
10.
by mrtsepelev 10y ago
April 11 (today), I believe. At least this date is mentioned on apply.ycombinator.com
11.
by mrtsepelev 10y ago
1) Yes, the problem with creating variations is still here and is worth mentioning. Today you need to know your product well and know your visitors well to create a valuable hypothesis. And this is what we're working on right now (disc
12.
by mrtsepelev 10y ago
The short answer is no. We're doing personalization instead of testing. It means that we're applying machine learning pipelines to deliver the best possible / most relevant content (page elements or landing pages) to every s
13.
Show HN: Landy – Machine Learning for Conversion Optimization(landy.io)
12 points by mrtsepelev 10y ago|2 comments
14.
by mrtsepelev 11y ago
In Russia same thing works in real-time (interpolating anything based on schedule will not be the best idea for busses / trams I believe) Mobile app from Yandex (works in Moscow & Saint-Petersburg): https://itunes.apple.
15.
by mrtsepelev 11y ago
As far as I know they didn't. But some government folks made a request to make this possible