Software Engineer, Developer Infrastructure

WHY NUMERIC?

Every business relies on accounting. Yet most software in the space was built in the early 2000s — clunky, slow, and far behind the curve on AI. Today, accountants are facing a growing, unsolved data problem https://numeric.substack.com/p/accountings-data-problem as a consequence of complexity, data, and system fragmentation — and they lack the tools to solve it. The problems are real, hard, and increasingly urgent. But they’re solvable https://numeric.substack.com/p/accounting-will-retrace-the-path.

Enter Numeric — the modern platform for accounting and finance. We're building the platform that realizes the dream of live, rich, trustworthy data and insights for companies to drive decision-making & execution.

We tackle foundational challenges required to automate ingest, transformation, matching, projection, and reporting for an end-to-end graph of financial data. Weaving together data, automation, workflows, and AI-first experiences, we’re empowering accounting teams to own the continuous window into companies' finances and activity.

We have strong product market fit and a growing base of customers who love our product — including Anthropic, Brex, OpenAI, Plaid, Wealthfront, and many more. We’re backed by top-tier investors including IVP, Menlo Ventures, Founders Fund, and 8VC, along with founders and executives from Ramp, Segment, and other category-defining companies.

About the role

Platform Engineering at Numeric is engineering for engineering. Our customers are our own engineers and, increasingly, the AI agents working alongside them. The mission: make every engineer and agent dramatically more productive by building a shared platform, so anyone can go from idea to safely deployed change without bespoke setup or tribal knowledge. This role is the developer-infrastructure half of that mission — environments, CI/CD and deployment, infrastructure-as-code, environment parity, observability, security, and internal tooling. The current reality has gaps (deploys with no staging gate, migrations without clean rollback, a prod-DB-clone lock that blocks deploys, Terraform drift, two clouds where there should be one) and a roadmap to close them. You’d own pieces of that roadmap end to end and then watch engineers actually use them.

What you might work on

- Deployment process: serialized, gated deploys across services; automated schema migrations with rollback; staging → production promotion that engineers trust.

- Environments & parity: local and remote dev environments that are reliable and fast to stand up; sandboxes and preview/review environments so reviewers can verify behavior, not just read code.

- Agent-First development: Including MCP gateways, governance, cost management, and foundation model selection.

- Infrastructure-as-code: Spacelift/OpenTofu, drift remediation, consolidating off DigitalOcean onto AWS, and IaC standards that are easy to follow by default.

- Repo standards & ownership: repository standards (linting, testing hierarchy, ownership manifests, idempotent onboarding), an ownership registry, and an internal developer portal.

- Observability foundations: local tracing/logging/metrics for every engineer, Datadog cost and hygiene, incident-management maturity, status pages.

- Security & access: unified secrets/config management, SSO in front of internal sites, IAM scoping, ECR scanning, audit logging, vulnerability management.

- AI Internal tools hosting: Numeric Labs — the platform that lets anyone at Numeric ship an internal tool without re-solving auth, hosting, and observability.

You might be a great fit if you…

- Are technically deep in infrastructure — AWS, containers/ECS, networking, IaC — and pragmatic about it.

- Have a product builder’s instinct for developer experience: sane defaults, tooling that teaches, “shipping means seeing it work successfully.”

- Like discovering the problem by meeting your users: pairing with engineers, watching onboarding, reading th …