About the Role
This is a full-stack software engineering role at an early-stage, venture-backed fintech company operating in the trust and fiduciary space. The company holds its own state trust charter, a regulatory asset that took real capital, approval, and ongoing examination to obtain, and it is live with clients today. The engineer who joins will design and build the core systems this company runs on: agentic workflows for fiduciary operations, a correctness-first fiduciary ledger, compliance integrations, and security-first infrastructure, all with close proximity to the founders and the clients themselves.
The ideal candidate has three to ten years of experience (up to fifteen for someone coming from a larger organization) building production backend systems at an early- or mid-stage startup (Series B or earlier). You are comfortable with a blank page: you turn user pain points into requirements and a working backlog without waiting to be directed, and you write code with AI tools every day as a matter of course. You bring a strong sense of ownership, a security mindset, and the patience to do accounting and compliance work correctly, not just quickly.
Key Responsibilities
- Design and build agent-driven workflows covering transaction monitoring, distribution proposals, document extraction and versioning, compliance checks (OFAC, KYC/AML), and required beneficiary notices
- Own the full agentic evaluation and annotation loop, including golden test sets, evaluation hooks, agent harness improvements, and LLM evals in CI/CD
- Extend and maintain the fiduciary ledger, which handles income and principal accounting on top of double-entry accounting and must correctly model splits, mergers, and complex corporate events, with correctness enforced at the database and API layers
- Build and maintain integrations with payment rails, compliance providers, and custodians, making deliberate build-versus-buy decisions grounded in correctness and security
- Design security-first infrastructure: self-hosted auth, envelope encryption for sensitive data, full audit logs, and observability through Honeycomb and PostHog
- Contribute to frontend work on the existing design system and component library, making complex trust and legal structures genuinely simple to understand
- Sit in on client onboarding calls with the trust officer and founders, absorb pain points directly, and translate them into product requirements and a prioritized backlog
- Lead projects from initial design through production, and help shape an architecture the company will rely on for years
Required Skills & Experience
- 3 to 10 years of experience building production systems, primarily backend (up to 15 years for candidates from larger companies)
- Demonstrated track record of taking projects from blank page to production delivery at a Series B or earlier startup
- Comfortable owning multiple domains at once with minimal process or structure
- Turns user pain points into requirements and a ticket backlog independently, without waiting for direction
- Uses AI coding tools daily as a standard part of the workflow
- Strong relational database skills, particularly PostgreSQL (constraints, checks, correctness at the database layer)
- Stable work history with roughly one year or more per company
- Authorization to work in the United States without visa sponsorship
- Ability to work on-site in New York City five days per week
Preferred Skills
- Applied AI experience: agent harnesses, eval and annotation loops, or MCP servers
- Fintech background, for example experience with Stripe, Plaid, or similar payment and compliance infrastructure
- Deep DevOps skills and comfort owning infrastructure end to end
- Strong security mindset, particularly around encryption, auth, and audit logging
- Experience with Go (backend) or React/TypeScript (frontend), though solid Postgres skills matter more than either
- Background in trust, fiduciary, accounting, or other regulated financial domains