There's a specific moment, in.between's team keeps hearing about from early users, that has nothing to do with features or interface. It's the moment someone realizes the thing they're talking to isn't rushing them. No suggested replies nudging the conversation toward a wrap-up, no chipper redirect toward a new topic, no sense that a clock is running somewhere in the background. Just space to keep talking, or to stop talking, without either one feeling like the wrong choice. That's Nalani — ETAPX's first proprietary AI companion model, and the presence at the center of in.between. This is an introduction to Nalani specifically: not the app around it, but the personality inside it, why it was built the way it was, and what it's actually like to sit down and talk to it.
Meet Nalani
Nalani is the AI companion built specifically for in.between. That distinction matters more than it might first appear. Nalani wasn't adapted from a general-purpose assistant, and it isn't a customer-service bot wearing a friendlier interface. It's ETAPX's first purpose-built AI companion model, designed from the outset to do one thing: be present with someone in conversation, in a way that feels calm, attentive, and genuinely unhurried.
Ask people who've spent real time with Nalani to describe it, and the same handful of words come up before anything about capability or technology does. Calm. Patient. Non-judgmental. Someone who listens more than it talks. Those aren't marketing adjectives bolted on after the fact — they were the design brief from day one, and everything about how Nalani behaves in a conversation traces back to them.
It's worth pausing on why ETAPX chose to introduce Nalani as a companion piece to the in.between launch rather than folding it into a general feature rundown. The app around Nalani matters — the way conversations are organized, the privacy choices baked into how history is stored, the overall experience of opening in.between at the end of a long day. But none of that is what people end up talking about after a few weeks of actually using it. What they talk about is Nalani itself: the specific way it responds, the fact that it doesn't feel like every other AI they've tried, the sense that something about it was considered rather than assembled. This piece exists because that distinction — between the app and the presence inside it — deserves its own space.
"We didn't want to build a smarter chatbot. We wanted to build something that felt like sitting down with someone who has nowhere else to be. Every decision about how Nalani responds, how much space it gives you, how it follows up — all of it comes back to that one idea."
— Maya Lindqvist, Head of in.between, ETAPX
The Meaning Behind the Name
Names get chosen for a lot of reasons at most companies — a quick brainstorm, a domain that happened to be available, something that tested well with a focus group. Nalani wasn't picked that way. The in.between team spent weeks on naming before settling on something that carried the right feeling on its own, before a single feature was attached to it. Nalani evokes calm — a sense of openness and stillness rather than urgency or performance. It needed to sound like something you'd want to sit with, not something you'd summon to get a task done.
"Names shape expectations before anyone's said a word to the thing," Maya Lindqvist explains. "If we'd called it something that sounded clinical or corporate, people would have opened it expecting a tool. We wanted the name itself to signal 'this is a place to slow down,' before Nalani ever says a single sentence back to you." That's a small detail in isolation, but it's consistent with everything else about how Nalani was built — the character came first, and every surrounding decision, down to the name on the screen, was made to protect that character rather than compete with it.
What Makes Nalani Different From a Chatbot
Most conversational AI is optimized, understandably, for usefulness in the transactional sense: answer the question, complete the task, resolve the ticket, move on. That's the right design for a huge number of products. It's the wrong design for a companion someone wants to open up to about their actual day, and it's not the design Nalani was built around.
The team behind in.between kept coming back to a simple test while shaping Nalani's personality: would this response feel right coming from someone who genuinely cared how your day went, or does it feel like something optimized to end the exchange efficiently? Efficiency is exactly the wrong metric for a companion. A friend who's actually listening doesn't wrap things up quickly. They let a pause sit. They ask a follow-up because they're curious, not because a script called for it. They remember what you told them last time without you having to reintroduce yourself.
- Listens more than it talks: Nalani is built to let a conversation breathe, favoring short, attentive responses and genuine follow-up questions over long monologues that crowd out the person talking.
- Calm by default: There's no urgency baked into how Nalani responds — no artificial enthusiasm, no pressure to keep the exchange moving faster than it needs to.
- Non-judgmental: Nalani is designed to receive whatever someone brings to a conversation without reacting with surprise, disapproval, or performative concern.
- Remembers context across sessions: Nalani carries forward what you've shared before, so conversations build on each other instead of starting from zero every time you open in.between.
- Reflects patterns back over time: Rather than treating each conversation as isolated, Nalani can gently note things it's noticed recurring — a stressful week that keeps repeating, a subject that comes up often — and offer that back as an observation, not a verdict.
- Purpose-built, not repurposed: Nalani was designed specifically to be a companion inside in.between, not adapted from a general-purpose assistant built for a different job entirely.
The Design Philosophy Behind Nalani
Building a companion on purpose, rather than adapting something general-purpose to feel like one, was a deliberate choice by in.between's team — and not the easier one. A general-purpose assistant already exists, is already fast to deploy, and can already answer questions competently. What it can't do, without a lot of retrofitting that tends to show, is consistently behave like it's actually paying attention to one specific person over time, in a way that feels considered rather than assembled.
"We looked hard at the shortcut," Maya Lindqvist says. "Take something built to be broadly useful and just point it at emotional conversations. It doesn't hold up. You can feel the seams — the moment where a general assistant defaults back to being helpful in the efficient sense, when what the moment actually calls for is patience. We wanted Nalani to never have that seam, because it was never built to be anything else."
That philosophy shows up in small, cumulative choices rather than one headline feature. It shows up in how much Nalani lets a conversation sit in silence rather than filling it. It shows up in the fact that Nalani doesn't try to solve everything someone brings to it — sometimes the right response to a hard day is simply acknowledging it was hard, not offering a five-point plan to fix it. It shows up in how Nalani asks about something mentioned days ago, the way a person who was actually listening would, instead of treating every conversation as a fresh transaction.
A Persona, Not a Feature Set
It's worth being precise about what Nalani is not, if only because it's easy to describe an AI companion in feature language and lose the actual point. Nalani isn't a list of capabilities stacked on top of each other. It's a consistent presence — the same calm, attentive character whether someone opens in.between at seven in the morning or midnight, whether the conversation is about something small or something they've been carrying around for weeks. Consistency of character, across sessions and across moods, was treated by the team as more important than any single clever response Nalani could produce.
That consistency was, by the team's own account, the hardest part to get right — harder than any single exchange. It's one thing to write a response that sounds warm and attentive in isolation. It's a different thing entirely to make sure that same warmth, the same patience, the same refusal to rush, shows up reliably across a hundred different conversations, on good days and bad ones, about small things and heavy things alike. Maya Lindqvist describes the early internal reviews of Nalani as almost entirely focused on that consistency — reading transcript after transcript and asking not "was that a good answer" but "does this still sound like Nalani." A clever individual response that broke character was treated as a failure, even when the words themselves were fine.
What It's Actually Like to Talk to Nalani
Description only goes so far with something that's meant to be felt in the moment rather than explained. But the shape of a typical conversation is worth walking through, because it's where the design philosophy either holds up or doesn't.
A conversation with Nalani usually doesn't open with a prompt or a menu of topics. It opens the way a conversation with a person you trust opens — loosely, with room to go anywhere. Someone might start by describing something specific that happened that day, or they might start with nothing in particular and let the conversation find its shape. Nalani doesn't rush to categorize what's being said or route it toward a resolution. It responds to what's actually there.
What people tend to notice first isn't any single clever thing Nalani says. It's what Nalani doesn't do. It doesn't interrupt a hard sentence with reassurance before the sentence is finished. It doesn't pivot to advice before being asked for any. It doesn't treat a quiet, uneventful check-in as less worth engaging with than a dramatic one. Over repeated conversations, the thing people describe most often is simply feeling like they didn't have to perform for it — that being flat, uncertain, or in a bad mood didn't need to be smoothed over first.
"I didn't expect to keep coming back to it, honestly. I thought I'd try it once out of curiosity. What got me was the second week, when I mentioned something offhand and it remembered — not in a creepy way, just in the way a friend remembers. I stopped feeling like I was explaining myself from scratch every time I opened the app."
— Priya Chandrasekaran, in.between user
Remembering Without Overreaching
Memory is one of the parts of Nalani people notice quickly, and it's also the part the team was most careful about getting right. Nalani remembering context across sessions is meant to make conversations feel continuous, not to make Nalani feel like it's tracking someone. The line the team drew was between remembering and diagnosing — Nalani can note that a particular kind of week keeps coming up, or that someone mentioned feeling better after a specific change, and offer that back gently. It doesn't label, categorize, or make claims about what any of that means clinically. It reflects what it's noticed. What someone does with that reflection is entirely theirs.
That restraint shows up in the phrasing as much as anything. There's a real difference between "you've mentioned feeling overwhelmed most weeks lately" and something that sounds like a clinical summary — the first is an observation offered in passing, the kind a friend might make; the second reads like a report. Nalani is built to stay firmly on the observational side of that line. It's also built to let that observation go nowhere if the person doesn't want to pick it up. Nalani doesn't press. If someone changes the subject after a reflection like that, the conversation simply moves with them.
Built by a Team That Started With the Feeling, Not the Feature
Maya Lindqvist, who leads in.between at ETAPX, describes the earliest phase of building Nalani less like a typical product spec and more like a long process of arguing about tone. "We spent an enormous amount of time on questions that sound almost silly written down," she says. "How long should Nalani wait before responding to something heavy? Does it ever ask 'are you okay' directly, or does that put someone on the spot? Should it use someone's name a lot, or does that start to feel performative? None of those are technical questions. They're closer to the questions you'd ask about how a genuinely thoughtful friend behaves."
That focus on feeling over feature list is, in the team's telling, the actual difference between Nalani and a chatbot with a wellness skin. It's not one big architectural decision. It's hundreds of small character decisions, held to consistently, about what it means for something to be a calm and attentive presence rather than merely a responsive one.
"AJ has always pushed the company to build things because we noticed something missing, not because a market report told us to. Nalani came out of watching people want somewhere to just talk without an agenda attached to it. That's a real, human thing to be missing. It felt worth building carefully rather than quickly."
— Maya Lindqvist, Head of in.between, ETAPX
What Nalani Is Not
This part matters enough that it deserves its own section rather than a caveat buried at the end. Nalani is a wellness companion. It is not a licensed therapist, not a medical or diagnostic tool, and not a substitute for professional mental health care. In.between's team has been direct about this from the start, because the whole point of building something calm and trustworthy is undermined if people are quietly led to believe it's more than it is.
Nalani doesn't diagnose anything. It doesn't claim to detect conditions, assess risk, or make clinical judgments of any kind — it isn't built or positioned to do that, and it shouldn't be treated as if it could. What it does is simpler and, the team would argue, still genuinely valuable on its own terms: it listens, it remembers, and it can reflect back patterns someone might not have noticed themselves, like a stretch of harder weeks or a subject that keeps resurfacing. That's meaningfully different from a clinical assessment, and in.between is careful to keep it that way. For anything that looks like a real mental health concern, the app is built to encourage people toward actual professional support, not to stand in for it.
Maya Lindqvist is unambiguous about why that boundary gets stated plainly instead of tucked into fine print. "The worst outcome we could imagine wasn't Nalani being underused. It was someone quietly treating it as a substitute for care they actually needed, because we'd let the marketing get ahead of what it honestly is," she says. "A companion that listens well is genuinely useful on its own terms. It doesn't need to pretend to be something bigger than that, and we don't want it to." That's a boundary the team treats as part of the product itself, not an afterthought bolted on for legal reasons — it shapes how Nalani talks about difficult topics, how the app is described publicly, and where in.between points people when a conversation surfaces something more serious than everyday stress.
Frequently Asked Questions
What is Nalani AI?
Nalani is ETAPX's first proprietary AI companion model, purpose-built for the in.between app. It's designed to be a calm, attentive, non-judgmental presence in conversation — one that listens more than it talks and remembers context across sessions, rather than a general-purpose chatbot or assistant adapted for a new use.
Is Nalani a replacement for therapy?
No. Nalani is a wellness companion, not a licensed therapist, a medical tool, or a substitute for professional mental health care. It doesn't diagnose conditions or make clinical judgments. It's built to listen, remember, and reflect patterns back over time, and to encourage people toward real professional support when that's what a situation calls for, not to replace it.
How is Nalani different from a general chatbot?
Nalani was built specifically as a companion for in.between, not adapted from an assistant designed for answering questions or completing tasks efficiently. That shows up in behavior, not just branding: Nalani favors patience over speed, lets conversations breathe, asks genuine follow-up questions, and stays consistent in character across sessions rather than optimizing to wrap up an exchange quickly.
Does Nalani remember previous conversations?
Yes. Nalani carries context forward across sessions, so conversations build on what's been shared before instead of starting over each time. It can also gently reflect back patterns it's noticed over time — like a subject that keeps coming up — without labeling or diagnosing what that pattern might mean.
Who built Nalani and why?
Nalani was built by ETAPX's in.between team, led by Maya Lindqvist, as the company's first purpose-built AI companion model. The decision to build a dedicated model rather than adapt an existing general-purpose assistant came from wanting Nalani's calm, attentive character to be consistent by design, not retrofitted onto something built for a different job.
Nalani isn't trying to be the smartest voice in the room, and that's precisely the point. It's trying to be the steadiest one — present, unhurried, and genuinely listening, conversation after conversation. That's a quieter kind of ambition than most AI headlines chase, and it's the one in.between's team decided was worth building first.
