HNHacker News
TopNewBestAskShowJobs

tonylucas

57 karma · joined December 24, 2008

Founder Workpiece.AI, Formerly: Head of CWM Tools @ Supermetrics, formerly Senior Product Director at Smartsheet (18-21) Co Founder of Converse AI (15-18, acquired by Smartsheet), Flexiant (09-14) XCalibre (97-09)

www.workpiece.ai

submissionscomments
tonylucas··on What happens when the Board of Directors begins to panic?
Posting at the direct suggestion of the author of the article (just to clarify I'm not trying to comment hijack), we (www.converse.ai) have solved a good deal of these problems, and are making it available as a platform for companies to build their own conversational services on.

We handle:

1) Front end integrations (Slack, Twitter, Intercom, Telegram etc, more to come)

2, Natural language interaction (either via our own NLP system or pluggable into other NLP systems as required)

3, Conversation state and context management

4, Validation and parsing of conversation input based on NLP, Regex, or external functions (e.g calling another API) as required. (NLP isn't best in all scenarios)

5, Static and Dynamic 'Conversation Maps', allowing you to generate them on the fly as required.

Comments/Questions etc welcome :)

tonylucas··on Enabling the NLP Conversational Economy
Thanks for posting, if I can answer any questions, happy to :)
tonylucas··on Facebook M – The Anti-Turing Test
I would hope in the future that more and more services are actually directly available to be interacted with over one or more messaging networks, enabling easy, asynchronous communication.

Whether Comcast would want to make it that easy for people to cancel is entirely a different issue though :)

tonylucas··on Facebook M – The Anti-Turing Test
You are presuming that they will employ all of the people needed.

Why for example would they not turn this into a platform that easily added the AI benefits for external vendors and services, and just be the middleman collecting a cut in some way.

They are already starting to integrate vendors into Messenger directly after all.

tonylucas··on Facebook M – The Anti-Turing Test
I think that also depends on who is actually completing/fulfilling the task, in the DMV example its AI adding context to a task that can't be automated easily.

The AI can still do the requirements/information gathering and leaving the job for a person to do. For common issues (support triage, customer service issues) even this still has significant value.

Equally from a training point of view, it may simply be the person correcting/confirming the AI is right, and leaving it to get on with the process, rather than fallover to the human completely.

tonylucas··on Facebook M – The Anti-Turing Test
I think the chat logs will help, but it won't just be about that data.

The data around how users interact will also be important:

Do they prefer a back and forth conversation, or do they want to say everything in one go.

Do they want to start a conversation, drop it, come back to it several hours later, or do they like completing it in one go.

How do they handle switching back and forth between different contexts, if certain requests take time, or do users not switch context.

What data are they happy to share, and what are they not.

What are the typical response times that a user considers acceptable 5 seconds, 1 minute, 5 minutes, 60 minutes? Does it vary depending on scenario.

Is there particular services or information that there is a trend towards, for example local search requests, research/information, particular types of purchase etc.

We've just spent 6 months going through a very similar process to this, which has helped drive the development of our Converse platform, which allows people to build semi or fully automated conversational messaging services, so this is fairly closely related.

(From our point of view, the NLP data we gathered was useful, but it wasn't the most important part)

tonylucas··on Show HN: Easily build your own conversational services/bots
More information on how it works is available at http://blog.converse.ai - I wrote a series of posts describing it's key functionality. Would love to get some feedback :)
tonylucas··on Travel planning software: The most common bad startup idea (2012)
The pre-trip organisation bit is something We are working on as part of a long term goal of our new startup, Planning To (www.planning.to)

Being able to use the context of peoples existing plans and calendar/schedules to provide automated concierge type suggestions of what would suit them best in terms of travel (e.g flights, hotels etc)

Focused on B2B travel rather than vacations though for reasons others have put very well in here :)

To begin with though, we want to solve the problems inherent in calendaring/scheduling in the first place, and then build a technology platform from there that people can use to add context and automation to all sorts of use cases, travel being one of them.

Anyone interested in what we are doing would really appreciate feedback, good or bad :)

tonylucas··on Introducing Stripe UK
Silicon Valley Bank (www.svb.com) might well do
tonylucas··on Ask HN: slicehost-like providers outside the US?
I believe it was myself who posted on the other list. FlexiScale certainly has had it's issues this year, there is no doubt about that, however we are all extremely confident about the new release that will be coming out in Q1 of 2009, and would really like to get people to try it.

We will also be issuing a full detailed decscription at that point of the internal issues we faced, to explain how we've overcome them.

Anyone who wants to try it at that point is more than welcome to a no-catches free trial, so we can prove we are delivering on our promises.

Regards,

Tony Lucas CEO, FlexiScale

← PreviousPage 2 of 2