AI Business Automation Engineer

Box•Redwood City, CA

1d•Hybrid

About The Position

Box is seeking an AI Business Automation Engineer to embed within functional teams (Finance, Legal, People, GTM, Customer Success) to identify and reinvent business processes using an AI-first approach. This role involves working closely with Subject Matter Experts (SMEs) to understand their workflows, design, prototype, and deliver technical solutions that transform operations. The position is within the IT organization and adheres to infrastructure, data, automation, and information security standards. Responsibilities span the technology stack, from integrating AI agents into financial processes to redesigning legal intake workflows and building secure data pipelines for customer intelligence. The ideal candidate is energized by the intersection of business problems, modern AI, and pragmatic engineering, with a desire to deploy code into production.

Requirements

2-3+ years of hands-on engineering experience in IT or software engineering roles, with a track record of shipping production code that real users depend on.
Deep experience with agentic coding platforms like Claude Code, Cursor, and Codex, as well as experience building custom agents that leverage MCP servers and CLIs.
Working proficiency in Python (or equivalent in Go, TypeScript, Java) — you write clean, tested, maintainable code and are comfortable contributing to a real codebase, not just notebooks.
Demonstrated ability to translate business problems into technical specs and delivered software. You can write a clear design doc, build the thing, and explain it to a non-technical partner.
Working knowledge across several of the following, with depth in at least one: cloud infrastructure (GCP preferred), data engineering (SQL, warehouses, pipelines), integration patterns (REST, GraphQL, webhooks, event-driven, iPaaS like Workato/SnapLogic), automation tooling, and IAM.
Familiarity with information security fundamentals — authn/authz, encryption, secrets handling, and safe data practices — and a willingness to learn what it takes to build systems that pass an audit.
Hands-on experience applying modern AI to real problems: LLM APIs, prompt engineering, retrieval-augmented generation, or building at least one AI-driven workflow, prototype, or agent.
Experience with agent frameworks (LangGraph, OpenAI Agent SDK, Claude Agent SDK, etc.), MCP, or building tool-using systems at scale.
Strong product instincts and a bias for shipping. You know when to deliver a thin slice and learn versus when to invest in something more durable.
Excellent communication skills. You build trust with non-technical partners and write specs, docs, and updates that travel without you in the room.

Nice To Haves

We are an AI-first company. This means you approach your work with a growth mindset and find ways to leverage AI to help make faster, smarter decisions that will 10X your impact at Box.

Responsibilities

Partner directly with functional SMEs to map current-state processes, identify friction and waste, and co-design AI-first reinventions that materially change how the work gets done.
Translate business requirements into clear technical specifications, architecture diagrams, and delivery plans, then build the agentic solution end-to-end alongside teammates and senior engineers.
Prototype rapidly using Python and modern AI tooling (LLM APIs, RAG patterns, agent frameworks, workflow orchestration) to validate ideas before scaling them into production.
Build and maintain integrations between Box’s core SaaS systems (Box, Workday, Salesforce, NetSuite, Atlassian, BigQuery, etc.) using APIs, iPaaS platforms, and custom services.
Design and implement automations that retire manual work, replace brittle scripts, and unlock capacity for the business.
Own the data plumbing — pipelines, models, access patterns — that makes AI solutions trustworthy and operable.
Apply security, privacy, and compliance controls from day one: data classification, least-privilege access, secrets management, auditability, and responsible AI guardrails.
Operate what you build: monitoring, runbooks, incident response, and continuous improvement based on real usage.
Evangelize what’s possible. Bring SMEs along the journey so they become co-owners of the AI-first future, not passive recipients of it.