New York City · Full-time · In-person
koodos labs is building personal intelligence that works for the person it represents.
Today, many of the richest portraits of us sit inside the services we use. They learn what we love, what we avoid, what we choose, and what we might do next. But that understanding is fragmented, built for someone else’s purposes, and rarely accrues back to us.
We think it should.
Our mission is to put personal intelligence within your control: to help people bring together the context they choose, understand and refine what Shelf learns on their behalf, and decide how that context can be used.
You should join koodos labs for the people and our mission. We aspire to build the best team of the 2020s, a place where it’s good to be from.
We built Shelf to power this vision.
Millions of people choose to connect Shelf to services like Spotify, Netflix, Amazon, YouTube, Goodreads, and more. Shelf aggregates that activity on their behalf. Every rating, comment, correction, or addition helps refine the picture, and our proprietary models turn those signals into structured, queryable personal context.
Users and creators have already built hundreds of hyper-personal apps on Shelf that put this context to work. Long term, we want people to be able to choose which apps, services, and agents can use the context Shelf has assembled on their behalf.
We’re looking for an experienced backend engineer to build the systems that make Shelf reliable, fast, and useful at scale.
This role sits behind some of the hardest product problems in Shelf. External services are messy: data can arrive late, duplicate, change shape, require backfills, or disappear. Unlike a normal data platform, failures here can become something Shelf incorrectly believes about a real person. Provenance, freshness, idempotency, schema evolution, and correctness are product concerns, not housekeeping.
You’ll build the systems that aggregate user-authorized activity from many external services, transform it into useful context on the user’s behalf, support real-time product experiences, and let downstream systems use only the context a user chooses to make available.
In this role you will: