Typewise Nova - AI customer experience that builds and improves itself
by•
Nova is the AI Operator that builds and improves your AI customer experience team, no developer required.
Describe in plain language how customers should be handled and connect your tools. Nova builds the agents, tests them on past tickets and shows you what failed before anything goes live. Once running, it monitors quality and proposes fixes 24/7 for approval.
Agents resolve whole requests, from orders and refunds to plan changes, across email, chat and WhatsApp. Unresolved requests are free.
Replies
how do you keep tests on past tickets from overfitting to the old way the team handled them
@mattkilmer the tests don't run on past tickets but on simulated conversations (multiple variants). based on past tickets you (or Nova) define the test cases, which then contain a scenario and an expected outcome (with optional assertions of what tools must/must not be called). but then simulations create variants of this. also the results don't automatically adapt your AI specialists, but if tests fail Nova will analyze why, and will suggest adapting the instructions if that's the best course of action.
@davideberle that separation makes sense, do you version the expected outcome and tool assertions when the policy behind a test changes
@mattkilmer good question, right now you'd update the test and the versioning gets logged, but a visible versioning should come in the future
Congrats on the launch! Thoughts on humalike.ai? good or bad fit?
@mcarmonas hmm, it looks cool! I'd need to see proper product specs to understand: what is it exactly, how does it work, and then I can give you a more detailed answer :-)
Congrats on the launch!
I'm curious about the "unresolved requests are free" pricing model.
How do you define what counts as "resolved" vs "unresolved"?
For example, if the AI gives an answer but the customer isn't actually satisfied, does that still count as resolved and get charged?
Is there a quality threshold or customer feedback signal that determines this?
@marketer_kim yes, a couple of things: (a) the AI judges when a ticket is likely closed, (b) the ticket gets re-opened if the customer writes again, even if the customer switches channels (--> bad answer in chat, customer writes an email, system recognizes this), (c) customer can give a rating at the end of the conversation; (d) and then there's Nova, that can for example on a weekly basis analyze tickets and see if there were conversations that likely were unsatisfactory to the customer, and then see why that happened, and how it can get improved.
@davideberle Ohh that's a really detailed set up! I can see why you say unresolved ones are free. Thank you very much for the answer! Wish you success. :3
Love the idea of making advanced AI customer service accessible to smaller teams too. 🙌
@fletcher_oliver thanks! do you have a team in mind that could benefit right now, and that was left out before?
Happy launch! "Runs, tests, monitors and improves” is a pretty ambitious loop. What part of the operation still requires the most human oversight today? Good luck today!
@davitausberlin usually the set up, and that's where Nova helps a lot... the other thing is the ongoing monitoring, just checking is all on track (= no action needed but takes time), or actually seeing how to optimize and then test whether it's an actual improvement.
1-2 hours for something simple vs 2-4 weeks for enterprise is honestly a pretty believable range, not the fake 5 minutes for everything claim these launches usually make. Congrats on the launch!
@salvatoresng true! i mean even setting up OpenClaw takes a while, i mean you can install it in 5min but it's not useful then yet. Nova has the benefit it's more narrow than OpenClaw plus it comes with a lot of best practices we've seen across the industry.