- It bot can read screenshots developers share and respond accordingly (currently available in Discord integration).
- It employs custom agents that handle specific tasks for your support workflow, such as querying transaction status via an API.
- Optimal Retrieval: We go beyond what's generally available, i.e., similarity search. We employ an ensemble of vector retrieval methods to find the highest-quality answer to a question.
- Reference URLs in answers: To improve the credibility of the answers, Helix AI provides links to trusted external references and sources for the information.
- User feedback registration: Users can provide feedback on the answers they receive, rating them as helpful or not helpful, along with comments. This feedback is used to improve the knowledge and quality of responses continuously.
I like the way you broke down things. People at a16Z have some sort of a network effect playbook for different types of Network Effects. I was listening about this in one of their podcasts.
actually, I submitted this post and have nothing to do with Greylock.
I am an entrepreneur myself trying to grow my product. Hence, I think about virality all the time. The audience here is also interested in growth as is clear by the upvotes.
I dont quite agree with that statement. If thats the case, what stops Pepsi from taking over Coca Cola's marketshare? In fact in blind tests, its very hard to tell which is what b/w coke and pepsi.
But, Coke commands the marketshare? Why? I think its their branding.
Branding is the single biggest defense against any incumbent. Universally agreed, branding is a very strong defense.
They had sales of 150 Million. Merely six times? Consider that against some Bay Area startups valuation !!
Is it only me thinking that DSC were undervalued? I think that DSC is worth way more than 1 Billion. I mean, this product occupies a much larger mindshare than some of its much bigger competitors.
Unless there are operational or growth issues, I can't understand why only 1B??
We did it this way at our company and it was pretty successful.
But keep in mind that we were just looking to tap in the existing AI technologies out there and leverage them.
A problem in the author's piece is putting India and Pakistan at the same pedestal. Consider this:
(1) Till date, not a single person has been charged in India for being homsexual. Being a homosexual is a crime in India. Yes, its a colonial legacy and in spite of indian bureaucracy being slow, people in govt are working to get rid of it. The less we talk about Pakistan, better it is
(2) Indian metropolis have a vibrant LGBTQ community. Suffice it to say, that there have been many pride parades in different cities of india. Contrast this with Pakistan, where a pride parade organized by US embassy in 2011 sparked protests.
(3) Author's claim about Sedition charges on university students - There have been massive protests against the govt in India because of this. Govt made a mockery of themselves with this action (as is evident). All the guys charged with sedition are still expressing their dissent against the govt and more forcefully now.
(4) To any of you who have been to the countries of Indian subcontinent, I am sure you can see the difference.
When I checked last, Google API does not allow to identify specific faces. It can detect faces but that's it. Clarify or Microsoft do.
Pricing wise almost all are the same.
In my view, Watson is a complete no no..