SODAX SDKDigital asset flows you can build and deploy with AI.
Promoted
Typewise is an AI agent platform that gives every customer a first-class concierge across service, sales, delivery, billing and retention. Its agents remember context, communicate across channels and resolve requests end to end by taking action in the systems you already use. Typewise Nova runs, tests, monitors and improves the AI operation, while your team controls what it can do, what needs approval and when people take over.
This is the 7th launch from Typewise. View more
Typewise Nova
Launching today
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.
Free Options
Launch Team / Built With
Typewise
Hi Product Hunt 👋 David here, co-founder of Typewise.
Over a year ago, we launched our AI customer experience platform. Since then, we've implemented it for dozens of companies, from large enterprises to small tech firms. We've helped teams migrate from Zendesk and Intercom, or add AI customer service on top of systems such as Salesforce.
One thing became clear: AI is not something you simply switch on and forget. Getting complex use cases to work reliably takes setup, testing and continuous improvement. The knowledge, instructions and skills all need to evolve as the business changes.
Most customer service leaders are being asked to manage this for the first time. Large companies spend heavily on consultants and forward-deployed engineers. Smaller teams often cannot, so they either fall behind or settle for a basic RAG chatbot and assume that is as good as AI customer service gets.
That is why we built Nova. It is the AI operator behind your customer-facing agents, helping you set them up, improve their skills and keep everything working over time.
For Product Hunt, we're offering 1,000 free resolutions for your first three months.
Try it with a real use case. If anything is unclear or you get stuck, tell me in the comments. I'll be here all day and happy to help.
@davideberle what happens when Nova encounters a request it has never seen before ?
Typewise
@dipanshu_kushwaha5 basically the company (together with Nova during the onboarding) defines what knowledge and actions the AI shall have access to (e.g. read CRM, read/write Shopify orders, etc.), and then Nova builds the AI specialists for that (e.g. shopping concierge, order questions and returns, warranty claims). So that way the system can already deal with a wide range of requests, no matter whether it has 'seen' them before or not. If the customer asks something that's out of scope for the AI, then, depending how you want this handled, it will typically bring in a human to take over. Then, Nova will analyze those "hand-offs" and may suggest expanding existing or adding new AI specialists, so that those cases are dealt with in the future as well. But that's then again your decision whether you want that scope expansion.
Buycott
Congrats on a very cool (and valuable) product. The 15-minute setup sounds very cool, but the reason these tools need so much setup is because the real world is full of exceptions. And good solutons often depend on decisions that are not documented. So I'm curious after those first 15 minutes, roughly how much more work does a typical customer need to do before they're ready to trust Nova to handle tickets autonomously?
Typewise
@scott_kennedy fair point. this really depends on the size / complexity. even after the 15min, easier things such as informative requests, a shopping concierge ("what is the right bike for me" for a bike seller, for example), a sales assistant (qualifying prospects, updating CRM, booking meetings), could be up and running in 1-2 hours of total time, tested and deployed to chat/email/messaging. for enterprise projects we typically budget 2-4 weeks total, also due to the slower cycles of larger companies. how does your setup look like?
Triforce Todos
Congrats on Nova, @davideberle
do the agents actually remember stuff from past conversations or does it reset every time a customer comes back?
BTW, all the best team )
Typewise
@abod_rehman it does remember! but it does also treat each conversation fresh, so for example, as an online retailer, the AI will freshly check your order history or your current order, and not rely on potentially stale memory of the AI, but properly check the factual systems, otherwise you get chaos and reliability goes down a lot. So I'd then also have the AI write important information back into the CRM at the end of the conversation, so you have clean customer records and not just an agentic memory. Does that make sense?
@abod_rehman @davideberle the write back at the end is where i would want a trace. whatever summarises decides what counts as important, and the crm field carries no record of that choice. if each written field kept the transcript line it came from, you get the same auditability you already insist on for order data.
Typewise
@abod_rehman @rabnoor_s yes, a write action to the CRM would exactly do that!
@davideberle @abod_rehman congrats on the launch, its a great product, really like the idea of it. good luck to the team.
Typewise
@abod_rehman @nancy_brown6 thanks!
Congrats on the launch @Typewise
how long does it usually take to get Nova set up and actually resolving tickets on its own?
Typewise
@boyuan_deng1 you should be able to set this up within 15min; the Nova onboarding guides you through everything and you can connect your systems via MCP. But that depends on your set up of course, feel free to share the use case and setup (also via DM if better) and I can give you my thoughts! Best is just to try it out directly :-)
1,000 free resolutions is a great way to test it with a real use case! What type of company benefits most from Nova?
Typewise
@imogen_wallace usually when you have, or, for early startups when you expect to have, a certain amount of customer communication volume, usually from 100-500+ conversation per month. but industry doesn't matter, we work with retailers, travel, tech, tele-health, e-sports, luxury, FMCG, utilities, etc.
The continuous testing and improvement part caught my attention. Is this mostly automated?
Typewise
@olivia_bennett7 yes, Nova helps define end-to-end test scenarios where then AI runs various simulations, which ensures reliability over time. Continuous improvement is then on Nova's watch, depending on how the rituals are set up that could be daily/weekly/monthly... still, on changes that affect things such as your processes you'd want to be in the decision-maker loop and not have automated 'improvements' happening... right? :-)
AI support sounds easy until the edge cases start showing up😄 How does Nova handle those tricky cases?
Typewise
@croft_benjamin it's exactly NOT easy! that's why you either see simple chatbots only or companies needed to hire consultants and engineers to get the setup right. Nova takes that over, giving you this team that typically managed the AI manually. A bit like how OpenClaw works, which also sets up the AI for you, writes scripts, etc., and you don't need to have a clue of the background stuff.