Staff Software Engineer
Pubblicato il 09-10-2026 - ancora in Turbigo
We are looking for a Staff Engineer to define the architecture of an AI-native accounting platform built to replace legacy management software .
Si candidi tempestivamente! È previsto un elevato volume di candidati per il ruolo descritto di seguito, non aspetti a inviare il suo CV.
Equity 0.3 – 0.5%
Fluent Italian is a hard requirement : we work closely with Italian chartered accountants.
Who we are
We are a team of 5: the founding team (CEO, COO, CTO) and our first two employees (Marketing and Data Engineer). We have raised the funding we need to build and are targeting 7 people in the tech and product team by end of year.
You are the fourth technical hire and the first with an explicit architecture mandate: you define the foundations we will build on for the next 5 years.
The context
We are not building accounting software; we are building the deterministic half of an AI-native system. Agents propose, your services validate. Business invariants that hold across thousands of companies. Accounting operations that are architecturally impossible to get wrong.
As Staff Engineer you are the technical reference point for the team. You do not just write code; you set the rules: how we model the domain, how we design APIs to be consumed by autonomous agents, how we guarantee tenant isolation, how we capture every system decision for the audit trail.
-
- Accounting domain modeling: double-entry bookkeeping, multi-currency, immutable journal entries, accrual-basis fiscal state. Correctness is non-negotiable.
- Agent-first API design: structured responses,
explicit error boundaries, stable contracts that an LLM can consume reliably. Not generic REST: APIs designed to be reasoned about by a machine.
- Multi-tenant isolation at scale: zero context leakage across thousands of companies. RLS, credential scoping, tenant-aware middleware.
- Event sourcing and audit trail: every accounting action is immutable, reversible only through a counter-entry. The log is the source of truth.
- Observability for AI: capturing the full decision context of every agentic action: input, output, reasoning, system state at the time of the decision.
What we need (must have)
-
- 7+ years of backend engineering in production on complex systems
- Documented experience of architectural ownership (decisions you made that still hold today)
- Python, Go, Java or similar statically-typed languages (real mastery, not familiarity)
- Advanced PostgreSQL: transactions, constraints, indexing, query optimization, data modeling
- Fluent Italian (domain communication with accountants happens in Italian)
- Ability to convey architectural decisions to the team clearly, in writing and in conversation
Nice to have
-
- AI/LLM systems: inference, RAG, agent orchestration, evaluation frameworks
- Event sourcing and domain-driven design at the system level
- Experience in fintech, accounting or financial services (double-entry, reconciliation, audit)
- Knowledge of Italian regulation (SDI, AdE, XBRL, PSD2)
- AWS and Terraform, IaC ownership
- Multi-tenant architecture with RLS and credential scoping at scale
Why it is worth it
-
- Fourth xysqume technical hire on a team starting from scratch: your architectural choices will stay in the system for years
- Explicit architecture mandate: you do not "contribute to decisions", you make them
- Greenfield, zero legacy, CI/CD from day 1, daily production deploys
- Direct report to the CTO, daily collaboration on the most challenging domain in Italian software
- Real equity: 0.3–0.5% on a company that has already found product-market direction
- Team growing: you become the technical reference for the whole team
- Equity: 0.3% – 0.5% stock options
- Meal vouchers and fringe benefits
- Direct report to the CTO
- HR screening (10 min): expectations alignment
- Architecture review (async, ~2h): we assess how you reason through a real design problem, not an academic exercise
- Technical interview (1h, remote): deep dive on the architecture review and architectural discussion on the accounting domain
- Final interview (in-person, Milan): meet the team
#J-18808-Ljbffr