Careers

Two people have spent six months building Nysa. We're looking for the third.

One role is open. It's remote, it's for someone in the US or Europe, and the terms are written out further down this page in full, including the parts that were uncomfortable to write.

What we are building

An app that is trying to make itself unnecessary

Nysa reads a person’s real birth chart and then talks to them about what it finds, in the language of psychology rather than prophecy. How they attach. What they do when they are hurt. Why they keep choosing the same kind of person and calling it bad luck.

It remembers. The tenth conversation starts where the ninth ended, not from nothing, which is the thing almost nothing else in this category does.

And it's built, deliberately, to make someone need it less over time. Not more. That one decision rules out most of the tactics that work in consumer wellness, and it's the reason this job is interesting rather than routine. You can't grow this product by making people compulsive about it. You have to grow it by being genuinely useful and then saying so clearly, which is a much harder and much better problem.

The category

Astrology isn't a small market. It's a large one full of companies that stopped thinking.

This matters because you are being asked to take a risk, and the honest way to argue for a risk is with other people’s numbers rather than our own. So here are three companies, none of them us.

AstrotalkIndia

About $145M in revenue last year, up 85%, reportedly preparing an IPO.

You pay a human astrologer by the minute. It scales headcount, not margin, and nobody remembers you tomorrow.

Co-StarUnited States

20M+ downloads. Acquired by Midjourney in July 2026.

A daily push notification with a sharp voice. It forgets you the moment you close it, by design.

CHANIUnited States

Roughly $14M to $18M a year on subscriptions, with no venture money at all.

Every word written by hand. Proof that people will pay properly for depth, capped by the cost of a human writing it.

Read those three together and the shape of the opening is obvious. One of them scales humans. One of them scales content. One of them refuses to scale at all. Not one of them is building the thing that actually knows you and gets more useful the longer you stay, because that requires memory, and memory is expensive and slow and doesn't make a good push notification.

An AI lab buying an astrology app in July 2026 wasn't a curiosity. It was a starting gun. This window is open now and it won't be open in three years.

Where things stand

The product is built. The audience is not.

Six months of full-time work by two people. Shivam is the founder and wrote the product. Kiara is the founding designer and owns how every part of it looks, moves and sounds.

What that has produced isn't a prototype:

  • A real astronomical engine computing real charts, not templated sign content pulled from a table.
  • A scoring layer that ranks psychological patterns from the geometry of a chart using fixed rules, and then constrains what the language model is allowed to say. The AI is bounded by math rather than left to free-associate.
  • A private memory system per person that recalls what still matters and archives what has stopped mattering.
  • Forty-plus product surfaces: chat, spoken conversation, a relational mode that reads two people together, subscriptions and credits, and a free public birth chart calculator.
  • Running. Live. Not a deck.

What doesn't exist is anyone whose job is to make the world aware of any of it. We haven't launched properly, there are no traffic numbers worth quoting to you, and we're not going to invent any. You'd be arriving at the start. That's either the appeal or the dealbreaker, and it's worth knowing which before you write to us.

How we work

Five things that are true about working here

These aren't aspirations pinned to a wall. They're how a two-person company already behaves, written down before it gets big enough to forget.

01

You own your function

We bring people in to own things, not to be managed. If you have to check with someone before you can act, we've built the wrong company.

02

Written, not scheduled

Decisions live in documents that anyone can read a year later. Meetings are the exception, and the ones we have are short and have a reason.

03

We don't count hours

No green dots, no status theater, nobody proving they were online at nine. We count whether the thing grew.

04

The numbers are open

Cap table, revenue, runway, what is in the bank. You're not betting your time on figures you're not allowed to see.

05

Direct, and kind about it

Respect means telling someone the truth early, while it's still cheap to fix. Both halves of that sentence are load-bearing.

The deal

There's no salary yet. We're not going to dress that up.

Every company we admire publishes what it pays. We can't publish a number we don't have, so we're publishing the whole structure instead, which is the same promise made a different way.

Founding equity, up to 5%
Vesting over four years with a one-year cliff, backdated to the day you actually start, so the months before the company is incorporated still count toward it.
A capped share of net revenue, from your first day
Not a bonus and not a maybe. While there's no salary to pay you, you take a defined slice of what comes in, which means if the work works you get paid before the company does.
A written salary trigger
On the first of a funding round or an agreed revenue milestone, this converts to a salaried role at market rate. That clause is in the agreement before you sign it. It's not a thing we tell you we intend to do.
15 to 20 hours a week
Keep your job. We're not asking anyone to give up their income for a year on someone else's conviction, and a person who still eats makes better decisions than one who doesn't.
All of it on paper, signed, before you start real work
Drawn up by a lawyer: the equity, the revenue share, the trigger, who decides what, and what each side keeps if you leave at six months. A handshake about equity is the most reliable way early teams destroy themselves.

Here's the arithmetic you're going to do anyway, so let's do it together. Nysa is a sole proprietorship in India today. It isn't incorporated yet, which means the equity is a written promise that converts on incorporation, not a share certificate anyone can hand you this week.

There is a real chance this is worth nothing. We think it's worth considerably more than nothing and the reasoning is all on this page, but the risk sits with you and no amount of good writing moves it somewhere else. What we can control is that every promise is written down and every number is visible to you from your first week.

Open role

One role, and it's the whole of go to market

Founding team

Founding Growth Lead

Nysa has a product and no audience. This role owns the entire distance between those two facts.

Remote, US or Europe15 to 20 hours a week to startView role

If that's not you but you think you should be here regardless, write anyway. A company this size doesn't have a headcount plan, it has a list of things nobody is doing, and a convincing email can add a line to it. shivam@intonysa.com.

How we hire

Three steps, and we reply either way

01

Write to us

shivam@intonysa.com. A few paragraphs beats a cover letter. Send one thing you've written that you'd defend, and tell us why this one.

02

A conversation

About an hour with Shivam. You get the cap table, the revenue, the runway, and a straight answer to anything you ask, including the questions that are awkward for us.

03

A small piece of real work, paid

We agree something real and narrow, and we pay for it. It's not a lot of money, and it's the first thing we'd spend cash on before revenue, because a week of working together tells both of us more than four interviews.

There's no deadline and no application form. We read everything that arrives and answer it, usually inside a week, and we close the role when we've found the person rather than when a calendar says to.

Before you ask

The five questions everyone asks, answered the way we would answer them on a call

Is this genuinely unpaid?

Yes, for now, and we'd rather you read that here than discover it on a call. There's no salary until we're funded or hit an agreed revenue milestone, and that trigger goes into the signed agreement before you start. What there is instead: founding equity of up to 5% vesting over four years, and a capped share of net revenue from your first day.

Why ask for 15 to 20 hours a week instead of full time?

Because asking someone in the US or Europe to give up a year of income on our conviction isn't a reasonable thing to ask, and the people who would say yes to it are usually not the people we want. Keep your job. Do this alongside it. It converts to full time with a salary when the money is there.

What happens if you never raise?

Then the revenue share is what you got, the equity converts to a stake in a small company that pays for itself or to nothing at all, and you keep everything you made. We'd tell you it wasn't working long before you worked it out yourself. That's the actual downside, and we're not going to describe it more gently than that.

Do I need to believe in astrology?

No, and a skeptic is arguably better at this. Nysa's positioning has to survive someone who does not buy any of it, which means the copy has to earn belief rather than assume it. Someone who needed convincing writes better for people who need convincing.

Can I do this from outside the US or Europe?

Not right now. The constraint is a two to three hour overlap window with the founder, and outside those regions the overlap stops working. Everything else here is async and written. If you're just outside and think it works anyway, say so in your email rather than assuming it disqualifies you.