Founding Engineer

a16z-speedrun•New York City, NY

3d•Onsite

About The Position

Omi Health is building a pet-health destination that combines proprietary dog longevity tests with user-reported and historical data to help pet parents extend their animals’ lives. The company aims to build a longitudinal data layer for the pet industry. As its most senior technical person, the Founding Engineer will set technical direction and own delivery end to end, making architectural decisions and helping shape the future engineering team. The role builds on a production stack that includes Python/FastAPI, Postgres and Supabase, a SwiftUI iOS app, React and TanStack web and admin surfaces, and AWS managed with Terraform. Key areas of ownership include designing a coherent animal health data model from fragmented sources; improving production LLM features with retrieval, evaluation, regression gates, failure analysis, and human handoffs; integrating unreliable external health-data systems; and establishing strong security and privacy practices. The engineer will also establish engineering culture, hiring standards, and the first engineering hires. In the first 90 days, the engineer is expected to improve the AI experience in the first week, bring a new data source live end to end by day 30, and build an initial learning loop for the longevity model by day 90. The role is intended for someone who can move from an ambiguous problem to a working production system without a detailed spec and who cares about animals. The company prefers commitment over procedures and would like the engineer to work in person; candidates seeking fully remote work are likely not a fit.

Requirements

Roughly 4–6 years of production engineering experience; the posting notes that relevant experience may be gained in less or more time.
Experience owning a backend service end to end in production.
Experience shipping an LLM feature and measurably improving it.
Experience integrating a challenging external system.
Strong Python skills and informed opinions about asynchronous programming.
Experience making an architectural decision that other engineers built on.
Ability to turn a vague problem statement into a working system without a specification.
Genuine care about animals.

Nice To Haves

Experience with multi-agent systems in production.
A background in healthcare.
Comfort shipping iOS or React features when the product needs it.

Responsibilities

Set the company’s technical direction and own technical delivery end to end.
Make architectural decisions for the product and infrastructure, and have a say in future engineering hires.
Design a scalable data model that consolidates fragmented and inconsistently coded animal health data into a coherent record for each animal.
Improve production LLM features through retrieval quality, evaluation harnesses, golden sets, regression gates in CI, failure analysis, and appropriate handoffs to humans.
Integrate external health-data sources with undocumented APIs, unreliable uptime, and contradictory payloads, using idempotency, reconciliation, and data-trust mechanisms.
Set and implement a security and privacy bar that goes beyond minimum legal requirements, and demonstrate it under diligence.
Establish engineering culture and hiring standards, and bring in the first engineers.
Improve the AI experience during the first week.
Bring a new data source live end to end by day 30, from raw payload to information a pet parent can read.
Build the first version of a learning loop for the longevity model by day 90, including what data is captured, what is learned, and how model improvements are demonstrated.