Software Engineer, Data Platform

San Francisco · Full-time · On-site · Engineering · Estimated base salary $125K–$225K, plus equity

About the company

Every business depends on accounting. Yet most software in the category was built in the early 2000s — slow, clunky, and years behind on AI. Accountants now face a growing, unsolved data problem created by complexity, volume, and fragmented systems, without the tools to address it. The problems are real, hard, and increasingly urgent. They are also solvable.

This company is building the modern platform for accounting and finance — the platform that delivers on the promise of live, rich, trustworthy data and insight companies can actually run decisions on.

The team takes on the foundational challenges required to automate ingest, transformation, matching, projection, and reporting across an end-to-end graph of financial data. By weaving together data, automation, workflows, and AI-first experiences, it gives accounting teams ownership of a continuous window into their company's finances and activity.

Product-market fit is strong, with a growing customer base that genuinely likes the product, including many of the best-known names in AI and fintech. Investors include top-tier venture firms alongside founders and executives from category-defining companies.

About the role

Underneath everything, the product is data: a customer's accounting history, their ability to pass an audit, their chance to catch mistakes before the auditors do. The Data Infrastructure team builds the financial data plane that makes all of that possible.

That plane is a lakehouse. It ingests raw data from ERPs, banks, payment processors, and customer warehouses, transforms it into typed canonical tables, and serves it to the accounting modules, the reporting engine, and the company's AI agents. The team is partway through a deliberate architectural bet — a ClickHouse-backed lakehouse, SQLMesh, and per-tenant isolation — and this layer is expected to absorb more new investment than any other this year. Every engineer on the team owns a slice of the architecture end to end.

What this person might work on

1. Ingestion at scale. Incremental, watermark-based extraction from API sources such as NetSuite, Stripe, Brex, Ramp, and HubSpot; file-based S3 inboxes; and Postgres-to-ClickHouse replication — with the goal of making every new source cheap to add
2. Data contracts and typed source data. Keeping downstream consumers away from untyped raw JSON by enforcing schemas at the boundary and making datasets easy to discover
3. Lakehouse schema management. A single declare → codegen → apply pipeline so schema intent is codified, reviewable, and applied consistently across every tenant
4. Data security. Scoping queries, building ACLs, and making certain customer trust is never compromised
5. Batch processing and event-driven accounting. Decoupling event processing from downstream failures and giving operators clear answers to what ran, what failed, and why
6. Data correctness and observability. Completeness checks grounded in platform data, ingestion health monitoring, two-phase re-ingest, and explanatory links tracing ledger entries back to raw events
7. Reporting scalability. The query patterns and materializations that let reporting drill from a board-level P&L down to a single transaction without buckling
8. Plenty more beyond that

You might be a strong fit if you

1. Have built or run data pipelines, lakehouses, or warehouses in production and hold real opinions on ETL, idempotency, watermarks, and backfills
2. Treat auditability and "fail early, fail explicitly" as design constraints rather than afterthoughts
3. Work comfortably across TypeScript services, SQL in ClickHouse and Postgres, and transformation tooling like SQLMesh or dbt
4. Like staying close to the people consuming your data — accounting module engineers, the reporting team, and AI agents — and shaping the interfaces they depend on
5. Want to understand how accountants think about source data, completeness, and close, and let that inform what gets built

Requirements

1. 3+ years of software engineering experience, with meaningful time on data-intensive systems
2. Strong SQL and a working grasp of columnar stores, partitioning, and query performance
3. Experience with a modern transformation or orchestration stack — SQLMesh, dbt, Dagster, Airflow, or similar
4. Strong general-purpose programming skills; the codebase is primarily TypeScript with some Python

Nice to have

1. ClickHouse, Iceberg, or Athena experience
2. Multi-tenant data isolation, PII handling, or SOC 2 and audit-driven controls
3. Prior exposure to financial or accounting data such as ERPs, general ledgers, or subledgers
4. Orchestration systems like Temporal or Inngest

Stack

TypeScript/Node, Postgres, ClickHouse (ClickPipes), SQLMesh, S3/Athena, Inngest, AWS

How the team works

This role is in-person in San Francisco.

Engineers learn the domain, meet their users, and exercise substantial creativity and agency over product direction. The team moves fast, operates with high integrity, and genuinely likes collaborating.

The company believes the best business is the one with the best individual contributors, and has built its culture to maximize their work. Great product and engineering are seen as the output of a desire to win, deep care for the craft, and pragmatism about shipping for overall impact. Engineers ground their work in a real understanding of the users, the domain, and the organization's goals.