Page 1 of 2

AI Engineer - Intern

Location: HSR Layout, Bangalore

About

Anyaa is here to eliminate the job of running the kitchen.

We are a young consumer startup with real households using the product every day. We’ve found our fit, and now we’re setting up our full kitchen.

Two founders, a product that works, and a kitchen full of ideas.

What we don’t have yet is you.

Role

We’re looking for an AI engineering intern who wants to build at the earliest stage of a company. You’ll work closely with the founders, have a say in what gets built, and own things from idea to production.

There are no layers between you and the customer.

What you’ll own

1. How Anyaa thinks. The agents that plan for real households, execute through everyday surprises, handle houses that have nothing in common effortlessly.

2. Evals. The evals and the infra that tell us whether a change made Anyaa better or worse.

3. Anyaa’s food knowledge. Our recipe and nutrition knowledge base, including regional Indian dishes and ingredients. You’ll grow it through real kitchen experiments, test it, and fix it where it falls short.

4. How Anyaa talks. Conversation personality on voice, text and every other interaction. Humans don’t like talking to AI. So we have to make sure Anyaa sounds and talks like humans.

5. Idea to execution. You won’t be handed a ticket. You’ll find a problem in production, design the fix, ship it, and check whether it helped.

Why food makes this interesting

A meal plan is not a recipe list. It is a family’s habits, a cook’s routine, a parent’s worry, and a long list of things nobody says out loud. A plan that looks right on paper gets ignored if it doesn’t fit the household. Building AI for this means learning how people actually eat and decide, and we care about that as much as we care about latency and accuracy.

You’re our person if you

- Build with AI coding tools every day, and have shipped something real with them that other people used

- Can tell when an LLM’s output is subtly wrong, in code or in production systems, and can explain why

- Know how to break a vague problem into a clear spec, because writing a precise brief is now most of the job

- Have opinions on how to judge AI output: what to measure, what to read by hand, and what a good test case looks like

- Debug by reading traces and asking what the model actually saw, not by guessing

- Understand how agents fail: wrong tool calls, lost context, confident answers with no basis, and quiet drift after a model update

- Write clearly, because prompts, specs and notes are your main tools

- Are curious about how Indian households cook, eat and decide, and test your work against that reality

- Work without waiting to be told what is next, and finish what you start

- Want to build from 0 to 1, not maintain something that already works

Nice to have

Experience with evals or LLM observability; Voice agents; a build you can walk us through, including what broke and what you changed.

Not for you if

You want to stay on a narrow model or research problem, or you hand off work that an AI produced without reading it first.

What you get

- Real ownership over AI that people use every day

- Direct work with both founders, who build alongside you

- A really hard problem to solve

- A real chance of a long-term place in the team

- A health-first team trying to change how India eats

How we hire

We move fast. If you’re excited about this, send us your resume and a link to something you built. Start October 20th. Or tomorrow.

Name

Email

Phone number

LinkedIn / Portfolio

This role is high-intensity 6 days a week, work-from-office role. Is that a work environment you’d genuinely enjoy?

A
B

Drop your resume