We're building an inbox that runs on autopilot. But are founders ready to hand over control?
Hey everyone,
We are gearing up to launch Phase 2 of the KortexMail beta, and I'd love a reality check from other founders and heavy email users before we finalize our onboarding UX.
The core premise of what we're building is an email client that constantly learns from your behavior over time to eliminate cognitive load. Specifically:
Behavioral Categorization: Instead of you setting up manual filter rules, it watches what you do. It learns which emails you reply to instantly, which ones you leave unread for days, and which newsletters you ignore, and silently starts organizing your inbox based on what actually matters to you.
Contextual Drafting: Instead of generic AI "smart replies," it learns your specific communication style and business context. When someone asks a common question, it drafts a reply exactly how you would write it, learning from your past interactions and corrections.
But here is my dilemma: As a builder, this sounds like the ultimate time-saver. But as a user, does this trigger the "loss of control" alarm?
For those of you drowning in email:
Would you actually trust a system to adapt to your style and silently manage your mail?
Or do you prefer the strict manual control of tools like Superhuman (where you make all the decisions, and hotkeys just help you execute them faster)?
Trying to figure out if people genuinely want to completely delegate their inbox to an AI, or if they just want better tools to manage it themselves. Would love your honest thoughts.
This is a really interesting trust problem.
My guess is that most people want automation, but only after the system earns the right to act on its own.
For email, I’d probably trust AI drafting before I trust AI silently organizing or hiding messages.
A good onboarding path might be:
So the product starts as an assistant, then becomes more autonomous over time.
The hard part is probably not the intelligence. It is making users feel like they can still understand and override what the system is doing.
i completely agree with you and thats our plan we will show user drafts and he can look at it and guide the agent and it will learn from the user because everyone has different way of writting.
KortexMail is solving a trust problem with a name that sounds like infrastructure.
That’s the mismatch.
For backend tooling, Kortex works.
For something reading, sorting, and drafting your email autonomously, trust gets decided before the feature set does.
This category is not just about automation.
It’s about control, privacy, and whether users feel safe handing over judgment.
That makes the name part of the product.
KortexMail sounds like the engine.
Not the thing people trust with their inbox.
Lyriso.com is a much stronger fit if the goal is to make autonomous email feel calmer, safer, and more user-trustworthy.
Great problem to tackle 👍
I think people want automation, but not full control loss.
Best approach:
→ start with suggestions (user approves)
→ then gradually automate as trust builds
Drafting will be easier to trust than auto-organizing.
Also, I’m running a small project (Tokyo Lore) where we test ideas like this with a focused group of builders.
Since you’re thinking deeply about this UX tradeoff, this could be a strong fit — happy to share more 👍