Member of Technical Staff (All Levels) - Agent Data

Basis AI•New York, NY

1d•Onsite

About The Position

Basis equips accountants with a team of AI agents to take on real workflows. We have hit product-market fit, have more demand than we can meet, and just raised $34m to scale at a speed that meets this moment. Built in New York City. Read more about Basis here. About the Team The Agent Data team ensures Basis agents have accurate data at low latency to do their jobs. About the Role As an Agent Data engineer at Basis you'll own projects completely from scoping to delivery. You'll be the Responsible Party (RP) for the systems you design. That means you decide how to build them, how to measure success, and when they're ready to ship. We trust you to manage yourself. You'll plan your own projects, work closely with your pod, and take full responsibility for execution and quality. You'll build systems that serve every part of Basis: AI, product, and internal agents. And you'll make those systems fast, reliable, and easy to understand.

Responsibilities

Build and standardize our data platform
Design data pipelines that ingest, validate, and transform accounting data into clean, reliable datasets.
Define schemas and data contracts that balance flexibility with correctness.
Build validation, lineage tracking, and drift detection into every pipeline.
Create interfaces that make data discoverable, computable, and observable throughout the system.
Translate accounting concepts into well-structured ontologies: entities, relationships, and rules.
Create abstractions that help AI systems reason safely about real-world constraints.
Design for clarity: make complex workflows understandable through schema, code, and documentation.
Own the architectural vision for your area and keep it consistent over time.
Run effective design reviews that challenge assumptions and drive alignment.
Mentor engineers on how to think about systems: from load testing to schema design to observability patterns.
Simplify aggressively—removing accidental complexity and enforcing clean, stable abstractions.