Senior Software Engineer – Agent Builder (Alwin by DataSnipper)

DataSnipper•New York, NY

3d•Hybrid

About The Position

DataSnipper is hiring a Senior Software Engineer for the Agent Builder team behind Alwin, its agentic automation platform for audit and finance. The team builds the product that enables firms to encode audit methodology into governed, versioned agents that can be run consistently across a firm. These agents perform multi-step procedures over client evidence and produce workpapers with numbers traced to their sources, while humans plan, review, and sign off. This hands-on role involves designing an authoring experience and agent-definition model, and building governance, publishing, evaluation, and production-monitoring capabilities. The engineer will own features end-to-end across a Python backend and React web application, work with related runtime and evaluation teams, and collaborate with product, design, and audit methodology experts. The work is open-ended, with an opportunity to help define how auditors author production agents in a regulated domain.

Requirements

At least 6 years of software engineering experience.
Strong, production-grade Python skills.
Hands-on experience building and operating production agentic systems, including control loops, tool calling, planning versus execution, retries, and fallbacks; experience limited to prompt-and-parse pipelines is not sufficient.
Experience with LLM APIs and agent frameworks across more than one model provider.
Solid experience building data-driven web APIs on a major cloud platform.
Strong fundamentals in software architecture, system design, and data modelling.
Track record of owning product features end-to-end, including frontend work in React or a similar framework.
Experience with unit, integration, and end-to-end testing strategies and a quality-first mindset.
Proficiency with AI-assisted engineering; the team uses Claude Code.
Excellent communication skills and ability to work directly with auditors and product partners to define correctness.

Nice To Haves

Experience building low-code or configuration products, workflow designers, rule engines, or DSLs.
Experience with agent evaluation, including datasets, offline and online experiments, and failure classification.
Experience with versioning, permissions, and publishing in a multi-tenant platform.
Experience with durable workflow engines such as Temporal or production multi-agent orchestration.
Experience with document processing, including OCR, classification, or structured extraction.
Domain experience in audit methodology, accounting, or fintech.

Responsibilities

Build an authoring experience that encodes a firm's audit methodology into configurable, multi-step agents.
Design the agent-definition model, including steps, tools, inputs, thresholds, planning behavior, and human-in-the-loop gates.
Choose abstractions that let non-engineers build consistent agents without writing prompts from scratch.
Work with the Agent Runtime team to ensure authored agents compile onto the base agent, tools, and orchestration layer.
Own versioning, publishing, and governance in the Agent Library, including permissions to build, publish, and run agents.
Enable builders to test and evaluate agents before publishing in collaboration with the Agent Runtime and Evals teams.
Instrument agents in production and use traces to identify problems with agent definitions.
Implement security, access-control, and audit-trail requirements for regulated customers, aligned with OWASP GenAI.
Own features end-to-end in the Python stack through to the React web application.
Architect services with clear contracts for other Alwin teams.
Raise engineering standards through code reviews, mentorship, and well-documented technical decisions.
Break down epics and manage ambiguity with product, design, and audit methodology experts.