By using object storage (Think s3 and similar) and not replicating data for HA (Not needed if using s3) which is done by legacy systems like Elasticsearch and Splunk.
Okay fair enough. But as a developer it grinds my gears when I work hard on something and like a moth to flame I get asked to open source it (so others can copy my hard work)
Wow. When I was a kid I once asked my stepdad why people who live in super dangerous inner city neighborhoods didn’t move.
His answer (some people can’t afford to move) didn’t satisfy me at the time. I thought to myself “bus tickets cost fifty bucks”. As a child I didn’t understand the nuances of life, but I also think there’s some power in childish naïveté.
It may be perceived that showing the right advertisements or showing pictures of cats isn't mission critical, and even just a toy. However, when massive amounts of personal information is being collected and shared by such applications I would say they're pretty important.
> In our most successful projects we are able to reach 95% accuracy
Is this the 95% of issues that are common between customers that could be and are handled at scale with a FAQ/knowledgebase that is surfaced to a user automatically through prompts?
How does it actually do on the smaller percentage of questions that are complex?
And that 95% is on the most successful projects. But what does the distribution of accuracy look like across all projects? If 2 projects out of 1000 are 95% accuracy that doesn't mean much to me.
FAQ/knowledgebase are not enough to reach 95%. If that was the case then this problem would've been solved long ago with the advent of semantic search engines.
You also need the following:
1) General principles of instructions split across topics - essentially distilling a support training manual into a flow chart of prompts.
2) Algorithms to detect best past resolutions to create new knowledge-base material on the go.
3) Continuous retrain and feedback loops which we have written models to perform on behalf of customers.
And continuous R&D effort into how these systems can improve.
It seems like a very convoluted path, but if people are for some strange reason more apt to end up at the FAQ via “chatting” with a tool like this rather than just looking it up directly, this still seems useful, right? I mean what’s the alternative, is there a “more self-directed users as a services” offering out there?
reply