Senior Software Engineer – Evidence Agents; Alwin

Audit and finance are still massively manual and we are changing that. Data Snipper is a $1B, bootstrapped unicorn with 600,000+ users across 180+ countries, already embedded in the daily workflows of top audit and accounting firms.

Now, we are taking things further with our Excel Agent, bringing AI directly into where the work actually happens. Unlike generic AI tools, we do not sit on the sidelines. Our AI operates inside Excel, with access to real documents and audit evidence, meaning it does not just generate answers, it does the work, with full traceability.

We are not just applying AI, we are redefining how audit gets done. If you want to build something category-defining at scale, this is the place.

We are looking for a Senior Software Engineer in our New York office to join the Evidence Request Management team behind Alwin
, our Agentic Automation Platform for audit and finance. Alwin agents run multi-step audit procedures over client evidence and return finished work papers with every number traced back to source. Humans plan, review and sign off.

Every audit starts with evidence, and getting it from the client is one of the slowest, most manual parts of the job. We are building agentic PBC ("prepared by client"): agents that draft requests from the audit plan, chase clients for what is missing, check what arrives against what was asked, and hand confirmed evidence straight to the rest of Alwin.

These agents run for days or weeks, react to events from people outside the firm, and must know when to act alone and when to ask a human.

This is a hands‑on role in a team that moves fast and holds a high bar. Your work goes in front of hundreds of thousands of audit and finance professionals at the world's largest firms, and their clients.

• Build agents that run the evidence request lifecycle: draft requests from the audit plan, assign, follow up, receive, review and close

• Design long-running, event‑driven agent behaviour that reacts to client uploads, replies and deadlines over days or weeks

• Build agents that validate received evidence against the request, flag missing samples or wrong documents, and decide when to escape to a human

• Work with the Agent Runtime and Document Intelligence teams on the tools, extraction and evaluation these agents depend on

• Model complex workflow state reliably across firm and client users: partial deliveries, re‑requests, rollovers between engagement periods

• Build the firm‑and‑client‑facing experience in React
  , including notifications, reminders and permissions for external users

• Connect requests to the rest of Alwin, so confirmed evidence flows straight into tasks, the workbench and downstream agent runs

• Implement security, access‑control and audit‑trail requirements expected by regulated customers and their clients, aligned with OWASP GenAI

• Own features end‑to‑end in our Python stack (
  FastAPI
  , Pydantic
  , MongoDB
  ) on Azure

• Instrument agents in production and use traces (
  Grafana
  , Open Telemetry
  , MLflow
  ) to drive reliability and accuracy

• Raise the engineering bar through code reviews, mentorship and well‑documented technical decisions

• Break down epics and manage ambiguity with product, design and audit domain experts

• 6+ years in software engineering, with strong, production‑grade Python

• Hands‑on experience building and operating agentic systems in production
  : control loops, tool calling, retries and fallbacks, human‑in‑the‑loop, not only prompt‑and‑parse pipelines

• Experience building workflow‑heavy products
  : multi‑step state, notifications, permissions, ideally with users outside your own organisation

• Solid experience building data‑driven web APIs …