• Design, develop, test, and maintain full-stack web applications. • Develop backend APIs and services using modern software engineering practices. • Build responsive frontend interfaces using contemporary JavaScript frameworks. • Utilize agentic AI development tools (e.g., Claude Code, Codex, GitHub Copilot, Cursor, or similar) to accelerate software development while maintaining high code quality. • Design and implement AI-powered workflows, automation, and intelligent agents. • Contribute to cloud-native application deployment and infrastructure. • Assist with containerization, CI/CD pipelines, and DevOps automation. • Support AI infrastructure, including model deployment, vector databases, retrieval systems, and GPU-enabled workloads. • Participate in software testing, quality assurance, and code reviews. • Collaborate with cybersecurity researchers to integrate AI into real-world security applications. • Document software designs, APIs, and engineering processes.

• Embed with a functional department for a defined engagement, running discovery to understand workflows and pain points before writing a line of code. • Scope, design, and build full-stack custom applications (not just automations or scripts) using modern frameworks, AI coding assistants, and internal platform APIs, moving from prototype to production quickly. • Own delivery end to end including architecture, build, testing, deployment, and the training and handoff that gets a department to actually adopt what you built. • Work hands-on in the code alongside AI coding tools (Claude Code, Codex, etc.) to compress build cycles, iterating with stakeholders in near real time rather than through long requirements documents. • Make explicit trade-offs between scope, speed, and quality as requirements shift, and communicate those trade-offs clearly to stakeholders and leadership. • Partner with IT, Security, and Data teams so every deployed application meets SentinelOne's standards for data security, compliance, and long-term supportability. • Codify recurring patterns, components, and playbooks from each engagement into reusable building blocks for the next department and for the broader engineering organization. • Feed field learnings back to platform, product, and data teams covering what worked, what broke, and what the next generation of internal tools should look like. • Measure and report success through production adoption and measurable business impact, not lines of code shipped or tickets closed.

• Embed with a functional department for a defined engagement, running discovery to understand workflows and pain points before writing a line of code. • Scope, design, and build full-stack custom applications (not just automations or scripts) using modern frameworks, AI coding assistants, and internal platform APIs, moving from prototype to production quickly. • Own delivery end to end including architecture, build, testing, deployment, and the training and handoff that gets a department to actually adopt what you built. • Work hands-on in the code alongside AI coding tools (Claude Code, Codex, etc.) to compress build cycles, iterating with stakeholders in near real time rather than through long requirements documents. • Make explicit trade-offs between scope, speed, and quality as requirements shift, and communicate those trade-offs clearly to stakeholders and leadership. • Partner with IT, Security, and Data teams so every deployed application meets SentinelOne's standards for data security, compliance, and long-term supportability. • Codify recurring patterns, components, and playbooks from each engagement into reusable building blocks for the next department and for the broader engineering organization. • Feed field learnings back to platform, product, and data teams covering what worked, what broke, and what the next generation of internal tools should look like. • Measure and report success through production adoption and measurable business impact, not lines of code shipped or tickets closed.

• Release the function's pipeline of production tools — prioritized by business impact as the discovery pipeline surfaces them. • Engage directly with business stakeholders alongside the Solutions Lead — sit in on discovery conversations, ask the technical questions that surface real constraints, and translate what you hear into the right solution form. The Solutions Lead opens the door; the engineer brings the technical eye that decides what actually gets built. • Build headless, API-first, and agent-callable by default — every tool is engineered so a human can invoke it directly and an AI agent can invoke it programmatically as part of a larger workflow. API-first design, structured I/O, clean tool contracts. • Pair-program with an AI-paired development environment (Claude Code, Cursor, Copilot, or equivalent) — treat AI-paired development as the baseline mode of work, not an enhancement. Sustained throughput is the load-bearing promise of this role. • Contribute to a pattern catalog that compounds — document reusable patterns, tool contracts, and architectural decisions so each solution makes the next one cheaper to build. • Build internal tooling for the function’s own operations — including, over time, a FinOps agent that monitors AI usage across the function and surfaces cost-optimization opportunities. • Engineer with token-economics in mind — model routing (Haiku for retrieval and classification, Sonnet for reasoning), prompt caching, output validation, retry/cost discipline. Cost-aware code is good code. • Partner with Internal IT on the graduation pipeline — work alongside IT to harden tools that prove themselves and ensure handoff readiness when a tool graduates to production-grade managed infrastructure. • Practice security-conscious AI engineering — secrets in Bitwarden, environment hygiene, awareness of data exposure risks, and adherence to internal security and AI Spend Finance policy controls. • Document workflows, decisions, and reusable patterns so the work compounds across the Finance AI Transformation function rather than living in one person's head.